I haven't seen comments regarding a big factor here:
It seems like OpenAI is trying to turn Sora into a social network - TikTok but AI.
The webapp is heavily geared towards consumption, with a feed as the entry point, liking and commenting for posts, and user profiles having a prominent role.
The creation aspect seems about as important as on Instagram, TikTok etc - easily available, but not the primary focus.
Generated videos are very short, with minimal controls. The only selectable option is picking between landscape and portrait mode.
There is no mention or attempt to move towards long form videos, storylines, advanced editing/controls/etc, like others in this space (eg Google Flow).
Seems like they want to turn this into AITok.
Edit: regarding accurate physics ... check out these two videos below...
To be fair, Veo fails miserably with those prompts also.
Not to be a downer, but even as someone very optimistic about technology and AI generally, "TikTok but AI" sounds like a societally terrible thing to try and create.
What's the benefit of this? Curious if anyone has a solid viewpoint steelmanning any positives they can think of.
As a social experiment to reveal how senseless and pointless pop entertainment could be.
(personal rant) I've been in a mild existential crisis since I read Amusing Ourselves to Death. Can one form of entertainment really be more well-regarded than another? Is fine art fundamentally different from pop art? Are there 'finer' pop cultures amongst all pop cultures? I do still think reading The Song of Ice and Fire is more meaningful than scrolling TikTok. The crisis part is that I can't justify this belief with words.
There are two completely distinct differences that jump out to me initially that I think may help justify your feelings:
1: Reading a long book demands focus on a longer timespan than scrolling TikTok, and with focusing on a single thing for a long time, we get a sense of accomplishment. I don’t know how to justify this as valuable, but for some reason I feel that it is.
2: The Song of Ice and Fire (and GoT) were consumed by a huge proportion of people, and you now have this in common with them. This act of consuming entertainment also grants you a way to connect with other humans - you have so much to talk about. Contrast that with an algorithmic feed, which is unique just for you - no one else sees your exact feed. Of course, there are tons of people that see some of the same snippets of content, if their interests overlap with yours, but it’s not nearly as universal as having read the same series of books (and there’s much less to talk about when you’ve seen the same 17-second short form video than when you’ve both invested dozens of hours in reading the same series of books).
I don’t think these thoughts fully justify your belief, but hopefully they provide some support to it.
I think the point 2 will rub many people the wrong way (me included) though. That would make reading Fourth Wing or Twilight a more connecting experience than most classics. (Nothing inherently wrong with that, but... you know...)
The classics were classic because they were the most available and the most popular stories of their time, and they meant more in an era where creating and disseminating media was difficult. I love to romanticize a world where we go back to the classics to connect with our past and present better, even if just for the sake of efficiency. For better or for worse media is more ephemeral which means getting to a common vocabulary is one step removed. It's really a fun time to be alive.
The thing is that literature, and art in general, should be more than just entertainment. It should edify the reader, communicate some concept, moral lesson or keen insight about the world.
Remember when you were taught to extract the "moral of the story" in school? That was the whole point. That form of communication is what makes art valuable and it definitely is what makes some art more valuable than others.
I've been watching these videos for about an hour now.
I really want to call this the "Suno moment" for AI video.
Prior to Sora 2, you had to prompt a lot of clips which you then edited together. You had to create a starting frame, maybe do some editing. Roll the dice a lot.
Veo 3 gave us the first glimpse of a complex ensemble clip with multiple actors talking in a typically social media or standup comedy fashion. But it was still just an ingredient for some larger composition, and it was missing a lot of the soul that a story with a beginning-middle-end structure has.
Sora 2 has some internal storytelling mechanic. I'm not sure what they did, but it understands narrative structure and puts videos into an arc. You see the characters change over the course of the video. They're not just animated Harry Potter portraits. They're alive. And they do things that change the world they're in.
Furthermore, Sora 2 has really good "taste" and "aesthetic", if that makes sense. It has good understanding of shot types, good compositions, good editing, good audio. It does music. It brings together so much complexity in choice and arranges them into a very good final output.
I'm actually quite blown away by this.
Just like Suno made AI music simple and easy - it handled lyrics, chorus, beat, medley, etc. - this model handles all of the ingredients of a 10 second video. It's stunning.
Sora 2 isn't the highest quality video model. It doesn't have the best animation. But it's the best content machine I've ever seen.
I can see this, it's extremely impressive from a technological standpoint, and I've already been caught by the first convincing fakes on Reddit (an army person giving an anti-Trump speech). But I'm also worried, as it's a super easy channel to create convincing fakes, mass produced 'content' for mass consumption, etc.
Now these things aren't new, fake videos / images go back decades if not a century. But they took some effort to make, whereas this technology makes it possible for it to take less effort than it took for me to write this comment.
Of course, it's always my choice; if I stop visiting Reddit and touch grass instead it really won't affect me directly.
Maybe some MAY end up in compiliations in ten years, much like Vines do today. But there will be a million times more tiktoks and a billion times more AI generated videos than there were vines, so if 0.01% of vines became memetic, the amount of AI generated ones will be infintesimal.
Content is all ephemeral on some time scale, but you can cache the near-term content to maximize the views and cut back on compute costs. Some model or human made it (the cost), it's trending (the value), so keep it around for a bit.
Everything has a relevancy and penetration decay curve.
The funny thing is, I think this law applied in the classical era (1950's, 1990's, etc.), we just weren't creating at scale to realize it.
Maybe it's just one dominant variable: novelty. I'd be curious to see how we might model this.
I think that short film is AI generated. I only watched like 30 seconds of an office scene in the middle but it spontaneously changed from daytime to nighttime with zero explanation.
He says it's not: https:/x.com/jasonjoyride/status/1973164183798816773
>> How do you get HD renders? im getting like super low res shit
>It's because this isn't AI
I haven't watched the film, but the premise is something about an orbiting space station. I could easily imagine scenes featuring rapid day/night cycles like astronauts experience on the ISS.
If you've read the classics, then you will likely find a circle you can connect too. I've gone through "The Malazan Book of the Fallen" and it's a signal to know who are truly in epic military fantasy.
> Can one form of entertainment really be more well-regarded than another? Is fine art fundamentally different from pop art?
It depends on what you want to get out of art.
Do you want human connection and shared cultural context so you can talk to real friends about things? Do you want virtual friends and connections? Do you want ideas to inspire you to create your own things, or change how you think?
Do you just want to distract yourself from how hungry you are, how much inequality is in the world, and how depressed you are, letting death draw closer?
All of those are valid things, and different art is more meaningful for different goals.
Scrolling tiktok fits into the last one, it's burning time to avoid thinking about things, moving you closer to death. Song of Ice and Fire builds a large coherent world, has bits of morality and human relation, and all of those can spark ideas and be related to your own human suffering, so it indeed feels more valid to me as a way to reflect and change how you think.
I think reading does force more long term focus, even if it's marginal for certain books. Certainly moreso than scrolling TikTok.
My personal process of grappling with this led to a focus on agency and intentionality when defining the difference.
Scrolling TikTok, much as scrolling Twitter or Facebook or Instagram or YouTube's recommendations would be, is an entirely passive activity. You sit back and you allow the Content to be fed to you.
Reading a book requires at least a bare minimum of selecting a book to read, choosing to finish that book, and intentionally choosing at any given time to spend your time reading that particular book. Similar things can be said for selecting movies. The important part in my mind is that you chose it, rather than letting someone or something else pick what they think you'll like.
The process of picking things yourself allows you to develop taste and understand what you like and dislike, mentally offloading that to someone or something else removes the opportunity to develop that capability.
I think there's arguments to be made against this view: how can you decide what to read or watch without getting recommendations or opinions? If you only engage with popular media isn't it just a slower process of the same issue?
But I do believe there is a fundamental difference between passivity and active evaluation of engagement as mental processes, and it's the exact reason why it is harder to do than scrolling is.
I think this is a pretty lazy dismissal as far as things go. Yes people "always complain" about many things, but that's the correct response to things that are always getting worse.
The gist of your linked article is that they were opposed to reading because they believed that reading distracted people from labor, which they considered undisciplined and immoral. Of course there also seems to be a healthy dose of misogyny associated with it:
> Poeckel's statement that women should acquire a certain amount of knowledge, but not too much, because then they could become a "burden on human society," is representative of many other texts in which reading regulations played a central role.
Then once you get to the progression of books > comics > movies > Youtube > TikTok (did I miss any?), you can observe a steady decrease in the amount of cognitive effort required to engage with the medium and a reduction in attention spans. Reduced attention span is a legitimate concern and it's only getting worse as time goes on (ask teachers).
I actually enjoy TikTok in moderation these days but I worry about people who struggle to engage with anything but TikTok, it's like a generational ratchet that only seems to go one way, towards shorter and shorter attention spans.
Maybe someone can make the argument that this won't actually matter, but it's incorrect to say that things haven't changed in observable and measurable ways, and that people are just complaining about nothing.
While I believe that long form content such as YouTube essays can actually be intellectually stimulating depending on how you engage with the video itself (e.g. do you actively watch it, or do you just have it on in the background?), I truly believe that 95% of TikTok is just mindless slop.
My S.O. probably spends 3 hours a day on TikTok/Reels and I seriously doubt they could remember even 10% of what they saw in that time. It's like a part of their brain turns off while scrolling.
The TikTok feed is an amalgamation of posts from lots of people who aren't collaborating. The Song of Ice and Fire is a single work by a single author (or so I assume). So it's more like you're reading a single humongous post that has been “liked” (bought, positively reviewed, critically appraised) by a shitton of people, compared to a firehose of morsels that barely anyone cares about.
I’ve been in an existential crisis after reading Postman and I’ve since reframed the whole dilemma thusly: one of the highest aspirations for a person is the act of creation, and the result one can often call art.
What is wrong is instead the routine consumption of art created by others in a stupor to rest from the drudgery of daily work.
Could it be the incentives? With regard to books, paintings, theater, etc. you have an incentive to produce something that is meaningful or at least entertaining. Generally the artist is attempting to turn abstract thoughts or ideas into something real or quantifiable.
But TV and Social Media have their incentives twisted. It's just about ads. They don't really care what you are seeing as long as you are seeing as many ads as possible. The joke about TV was that a valid description of it was advertisements with a little bit of entertainment sprinkled in throughout.
I'm not saying that people haven't been able to use these platforms to build anything meaningful, but that the incentives and the purpose of these platforms are not to entertain, but to keep you glued to the feed for as long as possible to see as many ads as possible (which is why I think "rage bait" is so common).
I think some forms of entertainment can have also redeeming qualities. A novel can be seen (also) as a form of entertainment but it can also be a vehicle for a message. The difference with social media sized alternatives is that with the latter the "consumer" is much more passive, at most it's expected to react emotionally without thinking. On the other hand with the former there is an interaction between the work and the reader/viewer. Some books have the ability to make you re-evaluate your beliefs and your values, without being manipulative. Art is not necessarily entertaining.
One analogy is to liken tiktok (and shortform content) as exploring the shallows. Walking around, close to the shoreline, you explore pieces of flotsam that the sea washes your way. You might spend a lifetime on this shore, walking up and down, but most would argue that you've actually never gone anywhere.
On the other hand, reading a book is like getting on a boat. You've made certain preparations for acquiring the vessel and set course through unknown territory. A journey away from the shore and away from what's immediately at hand, which can also turn out to be a journey towards self-discovery.
> The crisis part is that I can't justify this belief with words.
Here's one attempt; it's art versus content. Tiktok is content; it's people recording a video, sometimes in one take and publishing, sometimes in multiple takes with some editing etc, sometimes fully professional ones. But overall, it's cheap, rapidly produced content for cheap, rapid consumption. ASoIaF was a labor of years to produce not just a series of books, but a world, a rich history, and later on a multi-media enterprise that involved and employed millions of people, then entertained and excited hundreds of millions of people over the years.
AI is lowering the barrier to entry even more, with anyone able to just punch in some words - less even than this comment - and produce something. For someone to consume. Maybe one in a billion will be remembered or still popular in a decade (like how some of these cheap videos are still popular / remembered / quoted, think vines / memes). But the ratio just keeps getting worse.
ASoIaF to a TikTok video is like... ASoIaF to a tweet.
A good fiction novel has multiple aspects: of course entertainment/escapism, but also a larger point the author wants to explore. With Asoiaf George Martin wanted to break/subvert classic fantasy tropes. For example that the good guy wins (Rob Stark is marrying for love and punished for that in the red wedding) or the romantic knightly quest, here done by Brianne of Tarth, an ugly/strong woman instead of a male fighter.
I am not sure how important fiction novels are (compared to reading non-fiction books or biographies who tell true facts about the real world), but I would say they broaden the horizon of the reader? And there is a selection effect in that “literature” was done by pretty smart people.
Scrolling TikTok is often described as mindless and with people not describing later what videos they watched. In general short form content (TikTok, Instagram, X/Tweets) seem to be much more superficial than long form content (eg this hn discussion board).
> "[...] in its place, he identifies a different kind of crisis. Not the crisis of attention, but the crisis of interest."
Our attention in fact, has never been as fully absorbed as is today's. In place of books and architecture (as in the film), our attention has shifted towards more rapid forms. Yet in terms of hours spent, our 'attention' towards them has massively increased.
Is the crisis we're feeling then one of purported inattention, or a general loss of interest and satisfaction from our surrounds? What has spurred this crisis? Gabriel and Casey's conversation ends:
> "What about everyday life? Are we losing interest in everyday life?"
I prescribe Plato - Republic, book X, specifically. How can one set of shadows on the wall be better than another, as they’re just shadows, degraded representations of the real?
Or perhaps Aristotle’s Poetics - pop culture has value because it is mimetic, and AI generated pop culture is no less a mirror, just one which produces reflections of every moment, all the time - but rather than the grand catharsis we might experience in a work of literature with well wrought characters with whom we empathise, we find the void staring into us as we do into it. Hollow art for hollow men.
Like it or not, the void is culture, and has value because it reflects us, albeit through a glass, darkly.
One path that might help you work out your own personal justifications is to find two forms of entertainment you enjoy at near equal levels, but where one you view as valuable and another as a waste. Then look at how both impact your life and see if you can identify what makes one valuable and the other a waste. This not only gives you a good inside view of what is happening with both forms of entertainment, but removes any bias to see your own version as superior because both forms of entertainment belong to you.
I did this, found two things I did for fun, both consuming significant blocks of time. The one that felt useless left not real impact. I want to do more of it, but after spending hours on it, I'm no different than I was before (other than perhaps a bit more skilled at the form of entertainment).
The other form, which was the same thing from an outside perspective (for example, my parents would see them as the same) left me different. It led to me building new goals, reevaluating things happening around me, spend more time thinking about where I'll be in 10/20 years. It led to me walking an hour a day and to start jogging some to build my endurance, despite the form of entertainment being unrelated to physical activity. I don't think this is innately a property of one entertainment form over another, but more about my personal relationship to entertainment.
Using this, how do 'poorly regarded' entertainment impact those engaging in it, compared to 'well regarded' entertainment? Are their lives better for it?
Your comment makes me think that we have criminalized and squashed entertaining but obviously political writing out of existence :-) .
Say that somebody writes to make certain ideas more visible. For example, somebody wants people to buy the idea that amusing oneself to death is what we do (the book you mentioned). Somebody else perhaps has found that we are chronically depressed and cynic, when instead we should be thinking that a dead death itself is a fine trophy to hang on the wall during the march of progress[^1].
You can a) decide that you are set on your ways, thus entertainment should be pure and removed enough from reality so it doesn't mess with your deeply held beliefs and not read any of those books. or b) run the risk and read the thing with an open mind.
A lot of people are in the a) camp. Those who are in the b) camp would still like to be entertained a little.
There are probably ways you could explore this quantitatively by trying to measure the amount of novel latent information in the data you are ingesting, or by trying to quantify its effects on cognition.
Most short form content would probably score low. It’s short, for one, and it tends to be repetitive and lack anything like plot complexity or nuance.
Of course it’s not like trite pop is new. Way back in the dime store novel days it was called pulp. TikTok is just one of the latest iterations. People have always consumed dumb filler.
I read AOtD and I had a different takeaway: TV is great entertainment (watch all the trash you want!) but it's terrible for news and learning about the world. I found it pretty convincing--it's so clear to me that our societal discourse divebombed when TV news became dominant.
>The crisis part is that I can't justify this belief with words
Reading thousand page novels requires actively engaging with the material as you grow your vocabulary, and explore new ideas.
Scrolling TikTok on the other hand is a passive process. Could you recall even a quarter of all the videos you see on your TikTok feed in a single day? I would doubt it.
If society only consisted of the people in a given sector/industry, could it continue and flourish? If we only had engineers, how would society fare versus if we only had influencers? In this paradigm, there's no difference between fine art and pop art.
is a trash derivative "we have Lords of the Rings at home" wannabe, completely void of joy and feels like it's written by an angsty edgy teenager who hates the world and has learned about medieval history for the first time and wanted to add zombies and dragons, the most original fantasy tropes.
I would honestly and unsarcastically take a day of scrolling through TikTok over sitting through 1 chapter of ASoIaF.
You might enjoy some of my writings on formalizing meaning (see here[1] and follow the links). In some way, although not always reliable, you can say that if you feel A is more meaningful than B, that is already some kind of evidence for this assertion even if perhaps unreliable in some ways.
So there isn't necessarily some huge crisis that you need to justify: in some ways reality just is (and this includes subjective reality;).
Say if you ask why do the laws of physics conserve energy locally, you can actually argue that if it were otherwise actually life would be extremely more unlikely, as that tends to increase instability in various systems (both energy divergence and going to 0 makes life unlikely); but still I'm almost certain you could conceive of forms of life in non-energy-conservative systems (something like Conway's Game of Life, but maybe with more advance rules if you prefer). So while it makes sense that the physics in our universe is approximately locally conservative (maybe not exactly in GR?), in totality it's just kind of a brute fact, an experimental observation. Our theories help us devise say better experiments to test e. conservation, and in a way map out the landscape of consistent physical laws. But they don't tell you which realization of consistent or admissible laws you'll find yourself in.
Other way to phrase it, what you feel is in a way real. So if you feel in some fundamental way better reading A than B, then that simply reflects a property of reality and needs no further explanation. The only problem is that in some cases our judgement can be distorted, like by substances or maybe overwhelming blinding desires (that fail to reflect fundamental experiences) or by limitations of our memory, etc.. But if we assume this isn't the case (i.e. some pathological reason for your preference), then your feeling is valid irrespective of a wordy justification. I think some things really are subjective, but also believe in a fundamental and very complex way subjectivity is actually as objective as anything else. I think the fact that one experience is actually (with some important caveats and necessary context) better than another in what might be called essentially an objective sense, is one of the most counterintuitive things we will come to accept about the human mind. We tend to mistaken complexity (it's very complex to compare experiences) to impossibility (it's impossible to judge experiences objectively).
I believe in principle there might be the equivalent Laws of Physics (say Newtonian mechanics) for the human mind, but I suspect we're still very far from it, because it might require analyzing the network of n=100 trillion synapses in our brain. I think one day we might get there, but that would probably require something like a computational effort maybe at least several times n, or even on the order of n², or some other poly(n), and also poly(n) memory. If we think of one of the major objectives of physical law is to make predictions, and explain behavior, and say to aid in engineering and designing structures, I think one of the main objective of the laws of the mind would be say to predict whether say an experience or mental state is good or not, and explain why it is so; and then perhaps allow improving a little the design of things so that we have better experiences, that is, a better life. I guess this is already what say psychology, various spiritual traditions, philosophy and arts try to achieve (and I think gets already in many cases pretty close, maybe increasingly closer, to the still inaccessible extremely complicated reality of the human mind and brain).
Regardless, we often have to do our best with what we have today, which is our best-effort subjective judgement, aided by language various human disciplines :)
Steelmanning? It's infinite content, meaning our customers will never run out of new and interesting videos to watch, which will inspire them to feed prompts into our systems too and have it generate more videos. Money can be generated from multiple angles; we can charge a premium for generating videos beyond a small free tier once we've hooked the prompters, we can offer people or companies the option to promote their own videos so they get put into people's feeds, and we can insert generated ads from big corporate sponsors. It'll be lit.
Why should a commercial enterprise that has had billions of investments have benefits outside of earning money? Besides the entertainment value that the masses get from making and viewing these, of course.
Social media is also a wonderful tool for influencing participants and controlling them in the long term. In other words, behind the economic purposes there is a darker, more profound effect that is dangerous in the hands of a few powerful players. In the case of TikTok, that would be the Chinese state. Why shouldn't US Big Tech also be interested in this kind of power, in addition to the extra revenue?
Yup, which is also why various social media owners bent the knee to the administration, and now TikTok is about to become state-controlled too. The long term effects of subtle social media propaganda will become apparent in the years to come. Or, will be vocalized, they already are apparent - I'm convinced social media and related, 24/7 "news" media are a big factor in the right shift in politics worldwide.
I’m just sceptical that OpenAI would be making “TikTok for AI” if they really believed that we are on the verge of creating Artificial General Intelligence.
Is there? Creating AGI sounds like a great way to utterly upend every assumption that our economy and governments are built on. It would be incredibly destabilizing. That's not typically good for business. There's no telling who will profit once that genie is out of the bottle, or if profit will even continue to be a meaningful concept.
I hear this comment a lot and I don't get it. Let's say AGI exists but it costs $100/hr to operate and it has the intelligence of a good PhD student. Does that suddenly mean that the economy breaks down or will the goalposts shift to AGI being "economical" and that PhD level isn't good enough? I still haven't gotten a heard a clear definition of AGI which makes me think that it will break the world.
This is what Open AI themselves believe the risk is:
> By "defeat," I don't mean "subtly manipulate us" or "make us less informed" or something like that - I mean a literal "defeat" in the sense that we could all be killed, enslaved or forcibly contained.
It won't break the world, but it's warranted that it will break the world of people doing labor and getting paid for it. And when you think of it, even being a mediocre (or even moronic) investor is practicing a form of labor, so not even capital ownership is safe in the long run. And yes, generational wealth is a thing but there are tides that slowly shift wealth from A to B (e.g. from USA to China). Have a machine smart enough with even a sliver of motivation (intrinsic or extrinsic) to get some wealth for itself, and just watch what happens...
> "Come on," he droned, "I've been ordered to take you down to the bridge. Here I am, brain the size of a planet and they ask me to take you down to the bridge. Call that job satisfaction? 'Cos I don't."
Now I can see it can make for a fun party game, but that they seriously go after it, when their game should be leading models to do serious work ... is not a great sign to me.
Honestly if I learned anything over the past few decades it’s that I’m typically wrong about these kind of predictions, and society as a whole uses social media in a way that I just don’t comprehend. I would have never guessed a social media app whose biggest feature is “it disappears within 24h!” (even though you can easily screenshot everything) would become as big as it became.
Also, remember that it’s not about benefitting society as a whole, it’s about benefitting the investors. If the investors get rich at the cost of society, it’s a win for OpenAI.
Certainly, and that is the more pessimistic view that I have, that this is being developed with a view to introduce product sponsorships and advertisements.
I mainly am curious if anyone has the view that there is broader benefit to the development of this, after all, wasn't that the entire mission statement of OpenAI?
Quoting from their announcement on their site:
> OpenAI is a non-profit artificial intelligence research company. Our goal is to advance digital intelligence in the way that is most likely to benefit humanity as a whole, unconstrained by a need to generate financial return.
This feels like something constrained by the need to generate a financial return, and not something primarily focused on understanding physics and world models, to be blunt.
> I would have never guessed a social media app whose biggest feature is “it disappears within 24h!” (even though you can easily screenshot everything) would become as big as it became.
Or 'everything has to fit into 120 characters' (= Twitter). Or 'replies are designed to be maximally rage bait-y' (= Tumblr).
To be fair at least Twitter started with the SMS limitations, so it made sense to have the limitation in exchange for being able to update it with an SMS, when Whatsup was not so common.
People always like telling stories. Books, comic strips, movies, they're all just telling a story with a different amount of it left up to the viewer's imagination. Lowering the barrier to entry for this type of stuff is so cool.
I think you have to be pretty pessimistic to not just think it's really cool. You can find issues with it for sure, and maybe argue that those issues outweigh the benefit, but hard to say it's not going to be fun for some people.
>Books, comic strips, movies, they're all just telling a story with a different amount of it left up to the viewer's imagination. Lowering the barrier to entry for this type of stuff is so cool.
This response just never feels true to me. Many of the most successful web comics are crude drawings of just stick figures and text[1] with potentially a little color thrown in[2] and like half of the videos I see on TikTok are just a person talking into the forward facing camera of their phone. The barrier to entry in the pre-AI world isn't actually that high if you have something interesting to say. So when I see this argument about lowering the barrier to entry, I can't stop myself from thinking that maybe the problem is that these people have nothing interesting to say, but no one can admit that to themselves so they must blame it on the production values of their content which surely will be improved by AI.
I think people have a mistaken view of what makes some form of storytelling interesting. Perhaps this is my own bias, but something could be incredibly technically proficient or realistic and I could find it utterly uninteresting. This is because the interesting part is in what is unique about the perspective of the people creating it and ideas they want to express, in relation to their own viewpoint and background.
Like you pointed out, many famous and widely enjoyed pieces of media are extremely simple in their portrayal.
>Perhaps this is my own bias, but something could be incredibly technically proficient or realistic and I could find it utterly uninteresting. This is because the interesting part is in what is unique about the perspective of the people creating it and ideas they want to express, in relation to their own viewpoint and background.
I completely agree. And now that you mention this, I realize I didn't even point to the most obvious and famous examples of this sort of thing with artists like Picasso and Van Gogh.
If someone criticizes Picasso's or Van Gogh's lack of realism, they are completely missing the point of their work. They easily could have and occasional did go for a more photorealistic look, but that isn't what made them important artists. What set them apart was the ways they eschewed photorealism in order to communicate something deeper.
Similarly, creating art in their individual styles isn't interesting because it shifts the primary goal from communication to emulation. That is all AI art really is, attempts at imitation, and imitation without iteration just isn't interesting from an artistic or storytelling perspective.
This "barrier of entry" rhetoric reads like a pure buzzword dreamed up by AI pushers with no actual meaning to it. The barrier has NEVER been lower to produce books or comic strips or anything else like that. Hell, look at xkcd, there's nothing technically challenging about it, it's quite literally just stick figures, yet it's massively popular because it's clever and well thought out.
What exactly is this enabling, other than the mass generation of low quality, throwaway crap that exists solely to fatten up Altman's wallet some more?
What about the era of flash cartoons? Remember "End of Ze World"? In a way that's throwaway crap. Or it could have been written as a comic strip, or animated manually. But Flash kinda opened up this whole new world of games and animation. AI is doing the same.
One that comes to mind is a sort of podcast-style of two cats having a conversation, and in each "episode" there's some punchline where they end up laughing about some cat stereotype. Definitely low quality garbage, but I guess what I mean by "barrier of entry" (sorry for the buzzword), is just that this is going to enable a new generation of content, memes, whatever you want to call it.
> People always like telling stories. Books, comic strips, movies, they're all just telling a story with a different amount of it left up to the viewer's imagination.
It's not just different amounts, but different kinds. A (good) comic strip isn't just the full text of a book plus some pictures..
I think it’s really cool… and I’m still concerned about the long term implications of it. We’ve already seen a lot of TV get worse and worse (e.g. more reality tv) in a quest to reduce costs. It’s not difficult to imagine a reality where talented people can’t make great content because it’s cheaper to thump out bargain basement AI slop.
The democratization of storytelling is probably the best argument in favor, I'd agree. Thank you for the response!
I do find the actual generation of video very cool as a technical process. I would also say that I can find a lot of things cool or interesting that I think are also probably deleterious to society on the whole, and I worry about the possibility of slop feeds that are optimized to be as addictive as possible, and this seems like another step in that direction. Hopefully it won't be, but definitely something that worries me.
Personally, I think TikTok and other video platforms are already awash in AI. So, in my opinion, a platform that is explicitly declared as containing just AI videos is actually less disturbing to me. Every minute spent watching known-fictional AI videos on this platform is a minute not spent watching deceptive imagery disguised as reality on TikTok.
Please note that I’m not necessarily commenting on whether the existence of AI generated video is good or bad for our society, because I think it’s pretty moot what we think about it. It’s not going to just go away even if the majority of people here at HN or in general feel that it’s problematic.
Would be fun if this devolved into psychodelic, hypnotic videos that have no cognitively discernable content (like white noise) but evoke an urge to press the like button.
Half of TikTok is blurry videos wtere you can’t tell what’s happening overlaid with meaningless text and random noises, so this wouldn’t be much different.
Maybe finally this is the social media thing that will cause people on realize it is all too dumb, and get them to disengage? (Although I have thought this about every new development for the last ~15 years so I guess it is not too hopeful).
I’m surprised you can detect that TikTok but AI is societally terrible but are still optimistic about AI. Why is that? As a technical point, AI can already be used on TikTok, and every other platform, so whatever societal terror is coming from Sora 2 already exists due to AI.
It is a strange choice also because their model aims to be better than others, and obvious choice would be going after filmmaking market, like Veo did.
And in presentation they tell about social aspect and scrolling, and how they would limit the scrolling.
Are they confused and is it really just three guys working on it.
>What's the benefit of this? Curious if anyone has a solid viewpoint steelmanning any positives they can think of.
Revealed preferences. Keep giving the people exactly what they want (not what they claim to want), in unlimited quantities, until the message is received or we're all dead.
> We are giving users the tools and optionality to be in control of what they see on the feed
If this works it is a more powerful algorithm shaping mechanism than TikTok’s revealed preference feed. Even if Sora doesn’t take off, it could force TikTok to integrate something similar.
I’ve been predicting for years that the next stop for “social” media is a purely machine generated slop feed designed to keep people addicted. No human creators at all.
The question is whether people will eventually get bored with this stuff or if it actually will mesmerize people for huge fractions of their waking lives.
If the latter, I suspect we will outlaw it eventually. It’d be like legalizing hard opiates, literally, but minus the ODs and health damage.
I think a dedicated "TikTok but AI" is infinitely better than AI videos polluting other platforms. Of course, in practice, the latter is already the case, rendering the theoretical benefits of the former kind of moot.
Nonetheless, a platform for AI videos with an audience looking for them, rather than the horrible "boomer-slop" that is prevalent on other social media, is welcome in my eyes.
I like it. It's not a societal reform or anything. It's just people experimenting, like the Show HN section. There are some fun things in there and it's not as addictive as social media.
Instagram does not welcome this and I don't think they should. It is its own lane. And if it's just a place to sweep AI slop into, that's a good thing.
Isn't it kind of fundamentally better. A huge problem with tiktok is doing stupid things on video. On a site where the premise is fake video. Making stupid videos is a very different societal cost.
You're right that the societal cost is very different. I hadn't thought about people doing the stupid things on video, I think personally I focus on the effects of the consumption moreso.
Personally I think the problem with TikTok is largely based in hyperoptimized content specialized to your interest shaping your worldview and isolating your perspective of the world from others, as well as probably being pretty bad for the ability to maintain attention and engage with long form narratives and ideas. I don't really think TikTok is unique here, other than that it's the best in the game at doing it and keeping people's attention.
But overall I suppose I just see something like this as potentially worse in those regards, but maybe I'm overly pessimistic.
There will be a cohort of technically savvy youth who enjoy that all the fuddy duddies are self selecting themselves off the platform. There will quickly be a lot of fun memes and exclusionary references and lingo. It will be a hit. Just not with anyone over thirty.
My opinion is that unless there is some insane breakthrough in power efficiency with video generation, or if energy costs go down to zero, there is no way such a thing actually becomes profitable at the scale of scrolling TikTok. It is far more power efficient (and cheaper) to have people post their own content.
I think you're overestimating how much power LLMs consume. Let's say one video pegs a top of the line Blackwell chip at 100% utilization for 10 minutes. I think a Blackwell chip (plus cooling and other data center overhead) is somewhere around 3000 watts when running 100%. So that's about 0.5 kilowatt-hours. I suspect this is a severe overestimate because there's probably a significant amount of batching that happens that cuts down on amortized power usage, and non-pro Sora 2 might be processed with weaker, smaller models, but I'm not very confident.
Data centers seem to have wholesale rates of around 4 cents per kilowatt-hour on the higher end.
This gets you 2 cents per video. If you're generating 50 million videos per day (an estimate on the higher side of how many TikTok videos are uploaded every day), that costs you a million dollars a day.
So if you entirely subsidized for free the entirety of all of TikTok's video generation just using LLMs, I don't think energy generation costs exceed 365 million a year (and I think this might be very severely estimating costs, but there are some large error bars here).
I'm pretty sure OpenAI (or any company) would be pretty happy to pay 365 million dollars a year for the soft social power of something like TikTok. Just the influence this buys in politics and social discourse would be worth the pricetag alone.
And that's of course leaving aside any form of monetization whatsoever (where in reality you'd likely be charging the heaviest users the most).
N.B. I'm also not sure it's actually more power efficient for users to post their own content in absolute terms. It seems not unlikely that the amount of energy it takes to produce, edit, and process a TikTok video exceeds half a kilowatt-hour. But maybe you're focused solely on the video hoster.
Yeah I should clarify. This is a very vague estimate around "total energy spent for making a video you wouldn't otherwise do" which includes stuff like lighting, transportation, video transcoding on the server, script writing, actor coordination, etc. E.g. if someone drives somewhere solely to make a video they otherwise wouldn't, then it gets included.
I hedged as "not unlikely" because I'd need to think harder about the amortization of more energy expensive videos vs less energy expensive ones and how much energy you can actually attribute to a video vs the video solely being an activity that would be an add-on to something that would happen anyways.
But it's not just the energy expenditure of a phone.
(I also think that 0.5 kilowatt-hours is an overestimate of energy expenditure by potentially up to two orders of magnitude depending on how much batching is done, but my original comment did say 0.5 kWh).
Yeah that's fair (although the original comment was only talking about energy costs).
But this is kind of a worst case cost analysis. I fully expect that the average non-pro Sora 2 video has one to two orders of magnitude less GPU utilization than I listed here (because I think those video tokens are probably generated at a batch size of ~100 per batch).
I'm not sure you need a 'breakthrough', just many incremental improvements will do the trick.
We are also getting better at producing cheap power. For example thanks to intermittent sources like solar and wind, in many places electricity often becomes free in wholesale markets some times of the day.
AI generation (including video) currently takes at least a second, and users expect that delay. So that means inference is not latency sensitive and you can put these data centres anywhere in the world, wherever power is cheapest. Model training cares even less about latency.
At the moment, the hardware itself is too expensive (and nvidia has long backlogs), so people run them even when power is expensive. But you can easily imagine an alternative future where power never becomes cheaper than today (on average), but we have lots of AI data centres lying in wait around the world and only kicking into full gear when and where power is essentially free.
We are not getting better at producing cheaper power as the cost has increased per hour a lot over the last 50 years. But we are generating more power from different sources that are cleaner.
Power needs to be given away or people paid to take it is more of a function of limited storage abilities and limited ability to scale down rather then generating unlimited power. The free power is an issue with how the system is built (and the type of power source) rather than a sign of success. The same area has to import power at higher costs when the sun or wind isn't as powerful.
> Power needs to be given away or people paid to take it is more of a function of limited storage abilities and limited ability to scale down rather then generating unlimited power.
There's no need to scale down solar or wind power.
Yes, storage is another way to make money from power prices that differ over time.
> [...] the cost has increased per hour a lot over the last 50 years.
Some sources of power, like solar, have been dropping in price a lot recently.
There have been some big breakthroughs with hardware, though I'm on mobile and can't provide a link currently. I expect it to take done time to get into production though.
Also I suspect that this won't stay free very long. Silicon valley loves the model of starting free to build a user base and then monetizing more later
Many people are just playing with images and the distinctive styles that Midjourney (the model) seems to have developed. It's also trained by ratings and people's interactions.
When you make images you can dial down the "aesthetic".
This app might top the charts via hype initially but I can't see why someone would stick with it long-term compared to other alternatives. Plus creators would have to pay soon to make these videos, what are they getting back? Unless they can make money via this
After seeing the results of Sora 2, Meta and Midjourney are not yet competing at this level.
This is the "Suno" moment for video.
It's easy to make a really compelling composition. Something even Google Veo couldn't do.
It's not the best looking video model, but it has everything else -- rich editing, good voices and lipsync, music and lyrics, animation (cartoon, 3D, anime), SFX. It's wild.
The videos aren't single clips but rather complete beginning-middle-end stories that unfold over several cuts.
They don't have ads, it's paid members only. You can see other people's images, including the prompts, so it's an interesting way to learn what works, and to mutate prompts and images. There are many ways of recombining or breeding images.
They have an onboarding flow where you rate images and it tunes into your aesthetic preferences. You can create mood boards for specific projects.
So I would say it's more community than social media.
Are users of the $20 tier really going to have to deal with that obnoxious bouncing watermark I wonder? The previous watermark could be cropped, but I often didn't feel the need to as I use it for fun, but that would make me not want to show anyone.
When they launched Sora, one of the first things people did was rendering a person holding a cardboard with a message on it. It started by asking for features and eventually turned into people responding to each other.
One conversation I remember was complaining about people who constantly want AI pictures of anime feet.
> It seems like OpenAI is trying to turn Sora into a social network
No need to guess; In the article they say that the purpose:
We first started playing with this “upload yourself” feature several months ago on the Sora team, and we all had a blast with it. It kind of felt like a natural evolution of communication—from text messages to emojis to voice notes to this.
So today, we’re launching a new social iOS app just called “Sora,” powered by Sora 2. Inside the app, you can create, remix each other’s generations, discover new videos in a customizable Sora feed, and bring yourself or your friends in via cameos. With cameos, you can drop yourself straight into any Sora scene with remarkable fidelity after a short one-time video-and-audio recording in the app to verify your identity and capture your likeness.
Given that even absolute SOTA video gen models struggle with continual uncut shots longer than 60 seconds - positioning it as a Vine/Tiktok interface makes perfect sense. Turn your weakness into a strength.
I have been saying this for a long time--generative art is just fine grained consumption. Instead of searching through YouTube/TikTok for content that may interest you, you can now just ask your LLM to generate what you want to see that moment. It's the next evolution of keeping the masses addicted to their devices, and we as software engineers are gleefully supporting this ... because?
The other is possibly there’s no point in a thousand users all turning up to a blank prompt box and using a lot of resources to generate the same thing, or things they are not impressed by. A lot of users will ‘get what they came for’ initially just by seeing a bunch of good examples. Discussions around them will help them produce better outputs faster. Etc
Spend some time on the Sora feed[1] and you can see that a weird kind of social-creator network has sprung up around the service. Turning it into a social app makes sense in that regard.
Doesn't mean OpenAI can't do other stuff with it as well.
> It seems like OpenAI is trying to turn Sora into a social network - TikTok but AI.
Makes sense. I hate it, but the timing is probably good for them to try. There's going to be a mass exodus from TikTok in the US at some point, and those people will land somewhere.
As silly as I think all of these tools are, my eye tells me Sora is incrementally better at making short videos of red balls dropping into clear bowls. None of them are totally convincing, but the first Sora clip is easily the best of the bunch.
Is it easy to record a voiceover or add chosen audio? (Sorry I don't have an invite code so I can't try.) I could see some room for human jokes or short human-driven songs that could use a video backdrop.
Although we can tell they are inaccurate, what percentage of people can visualize the prompts better in their mind’s eyes? I bet a substantial number can’t even tell the clips are generated if posted without context.
In a few aspects, these world models are already pretty close to what we have in our brains.
People in this thread saying that this is the kind of content kids like should go on tiktok for a sec. This is not at all what young people watch, it's just bad content, and misunderstanding that feels out of touch.
I'm an adult. I personally found some of it funny and entertaining.
I like appointment television too. Sometimes A24 isn't pretentious enough for me. But I'm not beyond saying that there's absolutely a time and place for saccharine.
This content will grab eyeballs. I'll bet money on that.
It doesn't really matter what you or I think anyway. OpenAI is delivering a stream of hits and will continue growing while we debate on the sidelines.
I agree with the idea that they will like it, but I don't think it will look anything like this. I imagine native AI generations willl produce content will probably be instrutable to anyone older, requiring meme translations. Maybe a channel can be an AI decipherer. Hah.
I've long thought that AI will force new distribution methods because old media is so markedly against it... Maybe this is another Netflix vs Blockbuster moment.
This is why a group of single/childless men should not be behind products they think will be good for children. Children can watch Skibidi Toilet on loop a million times and not get bored, that doesn't mean it is good for them.
That seems like an awful use of technology like this. I would imagine they mean to use that for serving ads, but how do you even generate conversations with ai slop plus product placements? I could see it working sometimes but I doubt it scales.
I just asked GPT 5 to generate an image of as person. I then asked it to charge the color of their shirt. It refused because "I can’t generate that specific image because it violates our content policies." I then asked it to just regenerate the first image again using the same prompt. It replied "I know this has been frustrating. You’ve been really clear about what you want, and it feels like I’m blocking you for no reason. What’s happening on my side is that the image tool I was using to make the pictures you liked has been disabled, so even if I write the prompt exactly the way you want, I can’t actually send it off to generate a new image right now."
If I start a new chat it works.
I'm a Plus subscriber and didn't hit rate limits.
This video gen tool will probably be even more useless.
But if we find it drifts further and further from the truth in cases of biases in news articles, image generation and others we will find ourselves bombarded with historical deviances where everyone can be nudged to anything.
that's why the AI capabilities should be as decentralized and "localized" as possible - aka, i want to own the hardware and software for LLM, image generation, etc etc.
Until these ai capabilities are as neutral and un-discriminatory as electricity, centralized production means centralized control and policies. Imagine if you are not allowed to use your electricity to power some appliances, because the owner of the power-plant feels it's not conducive to their agenda.
They're struggling, it seems; AI can generate anything, but that includes stuff that goes against laws and morals, so they spend a lot of time to lock it down to avoid that, but people's creativity with prompts and escaping the safeguards knows no bounds. It's basically like the fight against spam, an endless game of whack-a-mole where usefulness fights with decency.
I get this all the time. Especially since GPT5, generating an image starts a massive chain where it confirms what you want, and asks you to say yes, and you say yes, and then it confirms again, and this can go on for 5-6 times. Then if you swear at it, it refuses to continue. It is insane. Fuck you, OpenAI
This is classic OpenAI heavy-handed censorship/filtering. Don't expect it to get any better; if anything, it'll get worse thanks to the "think of the children" types.
If you want an uncensored model that doesn't patronize you then your only recourse are local models, which, fortunately, are pretty good nowadays and are only getting better thanks to our Chinese friends constantly releasing a stream of freely-licensed models for everyone to use, unlike the "freedom loving" Western labs which don't release squat and make even Xi Jinping blush with how strongly they censor whatever they let us lowly plebs access through a paywalled API.
I can't help but see these technologies and think of Jeff Goldblum in Jurassic Park.
My boss sends me complete AI Workslop made with these tools and he goes "Look how wild this is! This is the future" or sends me a youtube video with less than a thousand views of a guy who created UGC with Telegram and point and click tools.
I don't ever think he ever takes a beat, looks at the end product, and asks himself, "who is this for? Who even wants this?", and that's aside from the fact that I still think there are so many obvious tells with this content that make you know right away that it is AI.
It's a fairly useful tool if you know how to use it. People will also play with it as a toy. It's much like the masses getting access to cheap video cameras and smartphones with good cameras. It's going to enable different content, it's not going to make more hollywood movies. This is an early example of what people will make: https://www.youtube.com/watch?v=jBwluRXtS2U . It's just one person making all of this on the side.
This was my reaction when I saw Meta’s “Vibes” app. Who wants to browse a stream of exclusively AI generated videos? Obviously Meta wants that because it’s a lot cheaper than actually paying real people to make content… but it’s slop.
This is not the final target. It's video generation now, but that's just a stepping stone. The real thing is that learning a generator is also learning a prior over videos, and hence over how the world works. The real application of this will be word models, vision-language action models, spatial AI and robotics. Basically a kind of learned simulator in which to plan and imagine possible futures, possible actions and affordances etc. Video models could become a spatial reasoning platform too. A recent paper by deepmind (using veo3) showed that video models can perform many high level vision tasks out of the box.
Don't think it's going to end here at some slop feed.
The final target of these "world models" on a 20 year horizon is entirely unmanned factories taking over the economy, and swarm of drones and robots fighting wars and policing citizens.
This is why hundreds of billions are poured into these things, cute Ghibli style videos and vacuum robots wouldn't be worth this much money otherwise.
What’s so romantic about working in factories? Automation and robotics will accelerate the economy the same way information technology did, and humans will work on better problems than performing repeated tasks on an assembly line or flipping burgers.
There are arguably more jobs today as a result of computers than there were before they were invented. So why is the assumption that AI will magically delete all jobs while discounting the fact that it will create careers we haven’t even thought of?
> humans will work on better problems than performing repeated tasks on an assembly line or flipping burgers.
Haha. The current wave of “careers we couldn’t think of” that tech companies have created include being Uber/Doordash/Amazon delivery drivers, data labelers for training AIs, moderator to prevent horrific content spreading on social networks,… with way weaker social benefits & protections than the blue collar jobs of old they replaced.
So yeah, I have a hard time buying this fantasy of everyone doing some magical fulfilling work while AI does all the ugly work, especially when every executive out there is plainly stating that their ideal outcome is replacing 90% of their workforce with AI.
With the way things are headed, AI will take over large economic niches, and humans will fill in at the edges doing the grimy things AI can’t do, with ever diminishing social mobility and safety nets while AI company executives become trillionaires.
I actually see robot food delivery services around me, so it might not even be long before those Doordash jobs get replaced by automation. Now I see neighbors starting to get drone deliveries from time to time. Starship used to deliver to the datacenter I used before (it was technically on a college campus but unaffiliated), and I had a coupon for free ice cream delivered through Wing the other day.
> So why is the assumption that AI will magically delete all jobs while discounting the fact that it will create careers we haven’t even thought of?
I think that in a vacuum you could reasonably believe that this might be the case but I feel like it isn't just about the technology these days, it's about the hunger c-suites and tech companies have for replacing workforce with ai and/or automation. It's quite clear that layoffs and mass adoption of AI/automation raises shareholder value so there is no incentive to create new jobs.
Will there be an organic shift away from Tech/IT/Computers into new fields? It might, but I think it's a bit naive to think that this will be proportionate to the careers AI will make redundant when there is such a big focus on eliminating as much jobs as possible in lieu of AI.
The hope is that we have no employment and we moved into a different form of society where AI takes care of us and allows us to focus on more spiritual meaningful things.
For now AI is deleting many of the jobs the computer created.
The reality is we will more likely end up in a society where wealth/power at the very top will grow and the masses will be controlled by AI.
more than controlled, enslaved - 24/7 location monitoring (but also no need to ever go anywhere, as everything will be delivered), "perfect" nutrition (fed via IV or tasteless shakes), only "intelligent" conversation taking place between you and an AI agent (even if initially resistant, AI will successfully convince you to drop ties to relatives, friends, that is if ever allowed to make friends), all news delivered via AI-curated channels but is meaningless anyway since AI can create fake video of any leader or important person committing crimes, lying, etc, also all evidence of YOU committing a crime, or just embarrassing stuff like having a sex drive will be recorded and used as blackmail. A "job" to keep you occupied much of the day but your output is never actually needed and discarded by your AI agent "boss".
How is this not entirely obvious to everyone that this is the future? Could be 20, 50, 100 years, but coming for sure.
There are no world models in there, it's trained on arbitrary images/sequences.
There are no world models in us, we learn from only specifics in topological space, stitched together in sharp wave ripples. Everything is from detached memories working through optic flow. That's not a world model, it's not even a model. It's an analog.
This whole world model thing is another branding phase after language models failed to deliver. After world models it will be neuro symbolic, then RL will sweep in like a final boss fight, and then... it still won't work. Notice anything about these names? They're walking pneumonia paradoxes.
The point is that video generation is not the goal in itself. Just like classifying photos as cat vs dog wasn't the goal in 2013. I know that Sora 2 is not a world model.
But what's coming is: Vision-language-action models and planning, spatial AI (SLAM with semantics and 3D reconstruction with interactability and affordance detection). Video diffusion models, photo-to-gaussian-splats, video-to-3D (e.g. from Hunyuan), the whole DUSt3R/VGGT line of works, V-JEPA 2 etc. Or if you want product names, Gemini Robotics 1.5, Genie 3, etc. The field is progressing incredibly fast. Humanoid robots are progressing fast. Robotic hands with haptic sensors are more dexterous than ever. It's starting to work. We are only seeing the first glimpses of course.
It's largely irrelevant in terms of intelligence. What you're describing is throwing out 2-D topological integrations (what we do to achieve optic flow ultra fast reaction times in motion), vicarious trial and error, and brute force imposing a machine wax fruit of motion dexterity. It's simply not analog to events the way we experience, it's been cooked up in cog-sci as imitation, but it's not even that. The more we understand the brain's architecture and process, the less relevant this gets, as it's not for legitimate long-term bio ware. There are no world models, the idea is oxymoronic as the topological bypasses this in scale invariance. It's all a dead end this binary, since eventually, analog will rule this with minimal energy and software and use an entirely different software. Think of any arriving too early industry, AI is irrelevant, the first step was reinventing software. It took the least efficient compute principle and drove it to irrelevance using machine vision as an endgame. The lack of redundancies is the tell.
The built environment fits the human form factor well. Imitation learning and intuitive teleoperation is also easier. But it won't be the only form factor. The quadruped form (like Spot) is also popular, as well as drones etc.
I think generally I agree with you that this is a stepping stone towards bigger/potentially more important things......but that doesn't change the fact that they've packaged it to consumers as something that seems like it has, at best, close to zero utility and at worst has incredible downsides. I'm not sure why releasing this to consumers helps achieve those goals.
Ad money to recoup the huge investments into datacenters that will do the training of the better models that do the things I mentioned. Meta is working hard on AR, glasses (project Aria), egocentric modeling and spatial AI. At some point they may also pull out the Metaverse idea too, they are still working on avatars too, it's just currently not so popularly hyped.
Try Jeff Goldblum in The Fly! I just re-watched and the computer he uses is scarily close to our experiences now with AI. In fact, the entire "accident" (I won't spoil it) is a result of the "AI" deciding what to do and getting it wildly wrong.
The main lesson I learned from the March ChatGPT image generation launch - which signed up 100 million new users in the first week - is that people love being able to generate images of their friends and family (and pets).
I expect the "cameo" feature is an attempt at capturing that viral magic a second time.
> The Sora 2 system card claims Sora can resist generations of "political persuasion".
To actually do that, it would need to have the evolving contextual knowledge of current events and the reasoning power to be able to identify prompts which, when requested, would likely have use in political persuasion, which would be a bigger breakthrough in AI than anything they are promoting.
Conclusion: it can’t actually meaningfully do that, though it will probably reject some subset of prompts involving topics pre-identified to it as politically sensitive.
This is a good point. It is hard not to see this tiktok video of an unblinking man smiling at a soft focus sasquatch and think of Marconi’s first broadcast across the Atlantic. I challenge any man to look at this unsettling video of a dog astronaut and not compare it in his mind to the first integrated circuit, or alternating current.
That's not what I'm saying, I'm saying many technologies enable a slide into fascism, and not having AI until a couple years ago didn't seem to solve the problem.
I like this reasoning. If a thing doesn’t exist, and doesn’t cause a problem, then when it does exist it won’t cause a problem. Like for example people died in wars before the invention of the machine gun, therefore machine guns didn’t really kill a significant number of people.
Your point was either that the fake video of a guy doing a backflip is the same thing as the invention of the radio or that if something bad is happening, nothing that happens can make it worse. Or somehow video generation software would have solved fascism in advance by previously not existing, which is proof that it is now good.
The first two are very smart but if you were talking plainly about the outcomes of time travel paradoxes I apologize for missing that
I wonder if they're going to license this to brands for heavily personalized advertisement. Imagine being able to see videos of yourself wearing clothes you're buying online before you actually place the order, instead of viewing them on a model.
If they got the generation "live" enough, imagine walking past a mirror in a department store and seeing yourself in different clothes.
At that point, why even buy the clothes? Influencers will just post the video of the mockup on social media, which is the only reason they were considering it in the first place. Save themselves the foot fungus.
Then take the next step. Why even spend money going out? Generate a video of yourself with fake friends at a party and post that, while eating ice cream alone at home.
I was criticising and making a joke prediction about the practice, not suggesting you actually do it.
I agree with you regarding online validation. I would even go so far as saying that depending on online validation or fame in general for happiness is unhealthy and anyone who does should make it a priority to find alternative sources.
We probably haven't even scratched the surface of what will be done with this tech. When video becomes "easy", "quick", "affordable", and "automatable" (something never before possible on any of those dimensions) - it enables countless new things to be done.
Its still just video though. Its not a new type of media. My guess is it will play out same as self publishing on amazon. Ultra specific generas that monitize the infinite long tail.
I feel like the main problem with buying clothes online is there is no way to tell if they are actually good or fit right. The photos are all fake where it's just an image projected on a stock photo of someone in a shirt. Doesn't tell you what the material is like, doesn't tell you if it actually fits (an AI video model is just making up the fit).
I don't have to imagine it because it's probably the most COMMON fantasy that people who work in advertisement and marketing have every day.
Now... take it a STEP further. Remember the scene in Futurama where Fry tries on the Lightspeed Briefs and looks in the mirror to see a rather aspirational version of himself?
its called Virtual Try On (VTO) and there are plenty of models going there for static gfx, it is very reasonable to expect soon emerge those for video VTO.
Accurate virtual try on however is quite difficult, and users will quickly learn to distrust platforms that just generate something that"looks right".
You can prompt with a normal size 8 dress and "kim jungle un wearing a dress" and it will show you something that doesn't help you understand whether that dress would fit or not. You can ask for a tube dress and it will usually give him a big bust to hold it up. It's not useful for the purpose of visualing fit.
It will definitely be used for such just like image models already are for cheap tenu clothes, and our onions shopping experience will get worse.
Maybe this needs purpose built models like vibe-net or maybe you cab train a general purpose model to do it, but if they were spending the effort necessary to do so they'd be calling it out.
You don't need generative AI for that at all, snapchat filters have existed for a decade and are the same concept. A lot of brands have already adopted that.
I'm surprised I had to ctrl-f this far for the first snapchat mention. Same. All I see here is snapchat except on any platform. Far from a tiktok competitor and far from revolutionary.
Or, on the genAI side, Google marketed this use case heavily for Flash Image 2.5 (even if that's not the same type of generative model because it's geared for editing, it's still in the taxonomy)
Seems like a nice feature but the most important aspect is "fit" and I wouldn't trust these models to do that accurately. They'll most likely make everything fit perfectly. Should be fixable tho.
When the dust settles , that's probably going to be the most common application of these video models. Making automated social content kind of defeats the purpose; people empathize with other people, not with AI . (I guess that's why they didn't also make their interview video via AI)
But Sora /VEO will probably also revolutionize movies and tv content
The most interesting thing by far is the ability to include video clips of people and products as a part of the prompt and then create a realistic video with that metadata. On the technical side, I'm guessing they've just trained the model to conditionally generate videos based on predetermined characters -- it's likely more of a data innovation than anything architectural. However, as a user, the feature is very cool and will likely make Sora 2 very useful commercially.
However, I still don't see how OpenAI beats Google in video generation. As this was likely a data innovation, Google can replicate and improve this with their ownership of YouTube. I'd be surprised if they didn't already have something like this internally.
> the ability to include video clips of people and products as a part of the prompt and then create a realistic video with that
This is something I would not like to see, I prefer product videos to be real, I am taking a risk with my money. If the product has hallucinated or unrealistic depiction it would be a kind of fraud.
They're really playing loose with copyright: you have to actively opt out for them to not use your IP in the generated videos [1]
Tangentially related: it's wild to me that people heading such consequential projects have so little life experience. It's all exuberance and shiny things, zero consideration of the impacts and consequences. First Meta with "Vibes", now this.
The example prompt "intense anime battle between a boy with a sword made of blue fire and an evil demon demon" is super clearly just replicating Blue Exorcist https://en.m.wikipedia.org/wiki/Blue_Exorcist
one of the example prompts is literally:
Prompt: in the style of a studio ghibli anime, a boy and his dog run up a grassy scenic mountain with gorgeous clouds, overlooking a village in the distant background
Wow that is dark, after Ghiblis staunch stance on AI.
These companies and their shareholders really are complete scum in my eyes, just like AI in miltech.
Not because the tech isn't super interesting but because they steal years of hard work and pain from actual artists with zero compensation - and then they brag about it in the most horrible way possible, with zero empathy.
Then comes losing the little humanity left the mainstream culture, exactly as Miyzaki said, leading to a dead cold and even more unjust society.
The Miyzaki quote is out of context, he isn't talking about generative AI but rather a 2016 animation of a creepy zombie whose limbs are controlled by AI.
While this is true, it's hard to imagine people spending years perfecting the style would be happy to see it copied effortlessly without any compensation while people who made the copying possible are rolling in cash.
This is not just about copyright infringement or plagiarism.
Automatically generating text, images and videos based on training data and a tiny prompt is fundamentally about taking someone's work and making money off of it without giving anything in return.
Don’t worry, I’m sure someone will roll up and claim that it’s just “democratization” of that style and the prompt authors exhibit as much creativity as the artists themselves.
Or they’ll claim it’s no different from a person looking at something and learning from it, implying that a multi-billion dollar company collating and labelling petabytes of data without permission to be used as the raw material to create their slop machine is no different from a human being being inspired by someone else’s art.
Luckily it doesn't actually copy the style at all.
No matter what text you put in the prompt you'll get /something/. Just because you put "studio ghibli anime" in the prompt doesn't mean you're going to actually get that out of it. It'll just be kind of yellow and blobby.
(Also, the style isn't from "people" but a specific guy named Yoshifumi Kondo who isn't around anymore.)
No, the zombie context is actually not that relevant, given he says "We as humans are losing faith in themselves" in response to the AI animation. He's clearly disgusted by the entire concept of machine generated art.
Also, he was calling them ableist because they said crawling was creepy but it reminded him of a disabled man he knew.
Though… I'm always surprised how respectful Westerners are about Miyazaki. Meanwhile you read other Japanese directors and they're saying all kinds of things about him.
In the full context, he is literally admonishing young developers who created ai and animation automation software as a possible alternative to handmade animation. He rips into them not only for their technical failure but for missing the point of what he does, which is human expression.
Kids are happy that homework takes less time. Teachers are happy that grading the generated homework takes less time. Programmers are happy they can write the same amount of code in less time. Graphic designers are happy they can get an SVG from a vague description immediately. Writers are happy they can generate filler from a few bullet points quickly.
But then someone comes along, notices people are not working most of the time, fires three quarters of them and demands 4x increased output from the rest. And they can do it because the "AI" is helping them.
Except they don't get paid any more. The company makes the same amount of money for less cost.
So where does the difference go? To the already rich who own the company and the product.
...the whole innovation enabling IT is based on a massive fraud or gaslight if you want - first having everyone to let go of their content (and un-own it blindly), then using it alongside everyone else's knowledge without consent to create a compressed blob of things which are then resold again.
Power creates more power, money creates more money.
Communism is tossing the frog into boiling water (tens millions of dead), capitalism is boiling it slowly (poor people in first world countries might not afford a dentist but they're not starving yet).
We need a system that rewards work - human time and competence.
There are really only 2 resources in the world - natural resources and human time. Everything else is built on top of those. And the people providing their time should be rewarded, not those who are in positions of power which allow them to extract value while not providing anything in return.
56 minutes, 4 downvotes, HN is truly full of temporarily embarrassed millionaires.
Does anybody here really think rich people deserve to just get richer faster than any working person can? Does anybody really believe that buying up homes and companies and raking in money for doing absolutely nothing is what we should be rewarding?
Homes are depreciating assets. You can't get rich by "buying up homes and doing nothing" because you'd lose money. Nobody is doing this, although a bunch of confused people on social media believe BlackRock is doing it for some reason.
I live in the most expensive housing market in the world.
That's because the value of the land under the houses is so high; the house itself is nothing special. But even then, it's mostly because of Prop 13, and it only works out if you live in the house yourself. There's still noone cornering the market in California houses. Almost all landlords only own 1-2 properties.
You get insurance to manage risk. You factor in roof repairs, vacancy rates when determining rent meanwhile your property value over the last 10 years in most places around the world have at least doubled and more.
Doesn't mean the next 10 years will see that growth but if you believe your country/area's population will grow it is probably a good investment for now in the western world.
Insurance is not free money and you can't simply distribute risks among your tenants because the risks are correlated.
Like I said, you can tell it doesn't work because these businesses don't exist. There are essentially no landlords who own multiple single family homes. They do exist for multifamily and commercial.
I know plenty of people who own multiple houses. And plenty of houses that get rented and paid to the owner/landlord.
Insurance is not free money its a pool of money gathered from monthly payments together to offset risk. You don't need to distribute the risk over all of your homes you buy a policy for each home.
This stuff is 101. And works all around the world. There is even an app called airbnb that will find short term rentals for your house.
As the other poster said, these risks can be mitigated in various ways. If the property appreciates 100% over five years, your costs associated with those risks are comparatively minimal.
They certainly don't constitute a depreciating asset!
That, and the dragon looking straight out of How to Train Your Dragon - I wonder if they have agreements with the right holders, or if they expect massive lawsuits to create free advertising for their launch.
Here is a direct example of a derived work, to the point where the prompt is "n orange-brown anthropomorphic dog sitting in a chair at a table in a room that is engulfed in flames, happy dog sitting on chair at a table viewed from the side, dog with a hat, room is burning with fire all across the room".
That's covered by Fair Use, I suppose they will argue this if they get sued. Interestingly, commons doesn't allow Fair Use, but the according to commons, "this is not a derived work".
Thank you, interesting!
I don't know that much about Fair use: if I understand well, the key is that the use should be "transformative", right? Am I correct in understanding that:
- if the original "This is fine" meme was under copyright, the dog picture would be exempted from copyright by Fair use as it's a transformation
- here it's not even needed since the original is not under copyright ("this is not a derived work")
It was a batshit insane decision, and a wrong one. Also: Commons doesn't allow for Fair Use images, so actually the decision was made that this wasn't transformative as it wasn't a derivative image.
You tell me if that was a derivative image or not. I argued it was, and the argument was completely ignored.
Kids will go to School V2 and have absolutely nothing in common to talk about because each one will have completely unique media entertainment at home.
It's certainly something that has people worried. I suspect a population decline and unbalanced demographics with too many old people per working age person are pretty much guaranteed. But personally I don't think it will be catastrophic civilisation level collapse. But at some point things will rebalance and turn the other way as resources, housing and such become more abundant.
I'm also hopeful we sort out the problems with big tech eventually. I was initially against it, but I'm starting to think Australia's plan to ban under 16s from social media is actually a very good idea.
This is already the case with the myriad of streaming services and choices of what people will let their kids watch or not. With my little kids, we tend to mostly watch PBS Kids content with a bit of Disney shows mixed in when it comes to screen time. We try to avoid seemingly empty hyper-stimulating content like Paw Patrol and others. But in the end a lot of the other kids in school/daycare talk about these shows and others, which can lead to the kids not having that kind of shared context. For instance, my four year old loves Wild Kratts, but practically nobody in his class knows the show. Meanwhile, he doesn't have any context for the various characters of Paw Patrol.
As a parent of pre and actual teens, this isn't "becoming", it is the de facto standard. Minecraft, Roblox, Fortnite are our kids' third place https://en.wikipedia.org/wiki/Third_place
But if the marketing fueling the industry is to be believed, every parent will be able to build a tailor-made game for their child. I know that won’t really how it’ll turn out but it’s a funny exercise to think about.
Assuming you have to generate new content for each viewer second watched yes it won't pencil out. But if you have a library of tons of content you can keep building out...
I've seen a lot of "this is impressive" but I'm not really seeing it. This looks to suffer from all the same continuity problems other AI videos suffer from.
What am I looking at that's super technically impressive here? The clips look nice, but from one cut to the next there's a lot of obvious differences (usually in the background, sometimes in the foreground).
>It’s fucking video made by a computer after you type a sentence
Geez. Breathe a little bit. It's always weird to see somebody who had zero involvement in the creation/engineering/design of a product so disproportionately defensive about it.
The concept of text to video since LTX, CogVideo, AnimateDiff is closing in on two years of development at this point so there's naturally going to be a little less breathless enthusiasm.
If you had experience with even the locally hostable stuff like Hunyuan, Wan 2.2, VACE, etc. you'd probably be less impressed as well. The video they demoed had more fast cuts than a Michael Bay movie - illustrating the exact problem that video models STILL suffer from - a failure to generate anything longer than 60 seconds. In fact I didn't even see anything longer than 10 seconds at most. Maybe it's tailormade for an ADHD audience who grew up on Vine Videos.
On a more positive note, the physics have definitely improved though you can tell in some of the shots that the coherency degrades the longer scene goes on (see the volleyball clip).
I'm assuming you've played more with AI video creation than I have.
Was there anything impressive here or is this mostly "meh"? I didn't see this solve any of the problems with AI videos but maybe this is solving something that I didn't know was a problem?
That's really what I'm trying to figure out with this announcement. Seeing 100s of comments about how impressive this is with no comments really discussing why has me trying to figure out what part of the hype I'm missing.
Good question. The one thing that Altman really seemed keen to play up was the whole integrate yourself into the video which from what I watched is definitely a step beyond the more conventional Image-To-Video models.
Depressingly that's probably a killer feature since if there's one thing people want to see more of it's themselves.
IMHO I also think the fact that they're trying to position themselves as a sort of infinite doom-scrolling tiktok lends support to the idea that their models are still only suitable for relatively short videos since coherency probably falls off a cliff after 30-60 seconds.
I think a lot of it is 'a dog trained to play the piano at a 4th grade level' effect: yes, it's really notable and impressive, but it gets old fast and isn't obviously useful beyond the novelty. You need to put in a lot more effort to get any actual value out of these tools and at that point their strengths and limitations become a lot more clear and it's obvious that they are not an everything machine.
I agree. But also I'd add that I'd be impressed and wowed if what you showed is "Actually, now the dog is playing piano at 5th grade level" or "Now the same dog can play the drums!".
Getting a "WTF are you talking about, it's a dog playing the piano, why aren't you impressed!" doesn't help the case. I've already seen a dog play the piano and this doesn't appear to be any better than the last time I saw it.
I'm not seeing comparisons of this to other video generation that already exists. That's what I want to see. Is this faster, more crisp, better at continuity? Why is this impressive vs competition?
I spend about 16 hours a week making AI videos. I'm not at all impressed by Sora 2. Open AI merely seems less embarrassing behind than they were, but still clearly behind.
As someone who loves seeing videos that appear after I type something, is there anything I should be noticing here? Or is this just bit of a special moment for people who feel, for some reason, there is some value for themselves personally in being deeply bought into OpenAI ecosystem and who aren't aware of the AI video scene as a whole?
I agree. It's really interesting that computers can make videos from sentences.
This, however, isn't the first AI capable of doing that correct? Didn't Sora 1, Veo, and others come out before this making videos from sentences?
Surely what makes this impressive isn't "This does what other people have done, including us Open AI".
Since you edited, I will to respond to your inflammatory edit
> Shocker lol
> Like a guy who hates tomatoes, bread, cheese, and pepperoni going, “This pizza sucks.”
This is a technology, don't reduce it down to preferences. There are obvious flaws with the video generating tech and one of the most annoying parts of talking to AI enthusiasts is the fact that they are unable to engage in honest dialog. All AI is amazing alien tech, it's always flawless and perfect.
What I've seen with AI video in the past is that they can make impressive on first glance looking videos but when you dig into them things are "off". Further, the continuity only lasts for 1 continuous shot. That creates videos where every 5 seconds you see a new shot and in the same setting things tend to drastically change.
Sora 2 appears to have all those problems, it doesn't appear to have solved any of them. That's why I ask "What's super impressive about this".
The same way I'd ask "What's super impressive about ChatGPT 5 vs 4". Snarkly saying "What are you talking about, it is a fucking realistic chat that can write a short story!" doesn't convince or impress me or anyone else that's a skeptic.
I'm not enthralled by AI. I'll happily use it when it makes sense and as it improves I'll probably use it more. For this, I don't see a significant improvement over the prior state of art.
> It seems like there was a 90% chance you weren’t impressed by these either…
And you'd be wrong in that assumption. When these first came out, particularly veo, I was quiet impressed with the photo realistic scenes it produced. Just like I was pretty impressed by "this is not a real human" website when it first launched.
> Yes, this is why I asked what % of comments you’ve made have been skeptical about AI. There’s nothing that will impress you.
New and novel things impress me. What's new and novel about this?
> I didn’t want to convince you I wanted to find out if I should have any interest in what you had to say.
Ah, so my opinion about stuff is only worth something if I said "I believe AI is the best thing that has ever been invented and I will kill myself to further it!!"
Go find a church if you can only talk with zealots.
They don’t seem to if 90% of what you have to say about AI is by your own admission at best tepid.
I honestly don’t understand how anyone can look at the state of AI today (even if the gains are perceived to be marginal compared to the last version) and be like, “meh, not impressive, better say something negative about it.”
But that seems to be the gospel here on the Church of HN.
These are hand selected from thousands of prompts. The person who doesn't use these tools thinks that you prompt the video like you are Scorsese directing a film. That is exactly what AI video is not.
More like you write a prompt and then try try try to get something that vaguely does what you wanted it to do. Most of the time though it doesn't even get close. The best video is usually stuff you didn't even intend. It is great for demo reals but that is about it.
They already made such a big deal about Sora 1 for months before it came out then what have you seen from it after it actually released? Nothing. Absolutely nothing.
Honestly, Midjourney blows away what I seen in this video as far as being pretty but Midjourney video is the same problem. Your imagination is filling in all these AI video features that don't exist.
I have used Sora many times! If you need to see a Scorcese movie as output in order to be impressed then you will be incredibly disappointed.
> It is great for demo reals but that is about it.
This is what is amazing! Imagine you go back to 2020 and say, “Hey I’ve got this website where you can type and it’ll return back a little computer generated video that looks pretty real.”
The only thing lamer than people talking about how unimpressive AI is are the people talking about AI’s environmental impact.
If you're already a heavy user of AI tools, you've seen or used previous generations already. So it's just a gradual improvement, nothing to get excited about.
Just like smartphones have been incredibly boring in the last 10 years because the only change has been "slightly more performance" or "marginally thinner".
Really impressive engineering work. The videos have gotten good enough that they can grab your attention and trigger a strong uncanny valley feeling.
I think OpenAI is actually doing a great job at easing people into these new technologies. It's not such a huge leap in capabilities that it's shocking, and it helps people acclimate for what's coming. This version is still limited but you can tell that in another generation or two it's going to break through some major capabilities threshold.
To give a comparison: in the LLM model space, the big capabilities threshold event for me came with the release of Gemini 2.5 Pro. The models before that were good in various ways, but that was the first model that felt truly magical.
From a creative perspective, it would be ideal if you could first generate a fixed set of assets, locations, and objects, which are then combined and used to bring multiple scenes to life while providing stronger continuity guarantees.
"open ai is so nice because they spoon feed us little pieces of dog shit every few days to acclimate us to swallowing huge quantities of dog shit every single hours of your life in the near future, praise our benevolent god Sam Altman", and you should cheer for it!
> I think OpenAI is actually doing a great job at easing people into these new technologies. It's not such a huge leap in capabilities that it's shocking, and it helps people acclimate for what's coming. This version is still limited but you can tell that in another generation or two it's going to break through some major capabilities threshold.
This is a truly _wild_ way to describe "this version isn't much better than the previous one". Would you say "Apple's latest iPhone is a pretty small marginal improvement over the previous one, but it's useful to help peopel to acclimate for what's coming".
Impressively high level of continuity. The only errors I could really call out are:
1/ 0m23s: The moon polo players begin with the red coat rider putting on a pair of gloves, but they are not wearing gloves in the left-vs-right charge-down.
2/ 1m05s: The dragon flies up the coast with the cliffs on one side, but then the close-up has the direction of flight reversed. Also, the person speaking seemingly has their back to the direction of flight. (And a stripy instead of plain shirt and a harness that wasn’t visible before.)
3/ 1m45s: The ducks aren't taking the right hand corner into the straightaway. They are heading into the wall.
I do wonder what the workflow will be for fixing any more challenging continuity errors.
The video was slam cut together to avoid continuity problems. There was a lot of fast camera motion and unconnected scenes.
Particularly bad was the snowmobile sequence. It was literally a different snowmobile in every cut.
The racing pool duck scene was a different pool in every shot.
About the only consistent thing was the faces that were spliced into the scenes.
I do not really see anything super significant in the demo. It looks like this suffers from all the same problems of AI generated video. They just hid it by avoiding more then 5 seconds in the same setting.
I’m wary of being that damning, this early. What I want to know is, should my video have these kinds of continuity errors, how easily can I fix them?
It’s ok for this to be a fun toy. (And fun toy while also being an astonishing piece of engineering.) But if it wants to push beyond fun toy then it would be interesting to see how that process works.
Will Sora2 help me sketch out a movie for me, doing 10% of the work where I have to reshoot the other 90% for real, or will it get me 90% there leaving me only 10% left to do “by hand”?
(This is the exact same question, I believe, which is being asked of the maintenance burden imposed by vibe coded products. They get you 90% then fail spectacularly leaving you having to do the bulk of the work again? Or they get you 90% of the way and you int have to fill in the gaps to reach a stable long term product?)
I don't see how this is usable for making like a feature film. Editing will be impossible. At best it will be for ads. At worst for making social media slop.
It is not even just the errors. These video models are really impressive as long as you don't actually have something in your head you want to make. Then the laughable limitations are on full display.
I will believe it when I see because Sora 1 is probably the most disappointing technology given what I thought it was going to be that I can even think of. I waited forever for it and then barely used it because it sucks.
Not sure if it counts as a continuity error, but in the example "Prompt: Martial artist doing a bo-staff kata waist-deep in a koi pond", his wooden staff changes shape several times, resembling a bow at points. That was the first example I noticed as "clearly AI."
It definitely seems like a state-of-the-art model, but the sound having an underwater effect is the biggest tell. How long can it keep a scene going without things falling apart?
State of the things with doom scrolling was already bad, add to it layoffs and replacing people with AI (just admit it, interns are struggling competing with Claude Code, Cursor and Codex)
What's coming next? Bunch of people, with lots of free time watching non-sense AI generated content?
I am genuinely curious, because I was and still excited about AI, until I saw how doom scrolling is getting worse
Of course primary cause could be ZIRP, but AI definitely accelerated the problem.
Interns at big tech maybe impacted less, because their systems are so complex, but when I look at job boards or talk with engineers I see they're mentioning interns less, AI assisted coding more.
Bar for the interns is higher now, why do I need 3 interns to polish the product if I can complete 70% of the job with AI and hire 1 intern to fix other parts
Not sure your reply warrants any further expenditure of effort on my part, but for the benefit of other readers:
The bootcamp (actually, evening classes in coding run in cooperation with the public sector) regularly placed graduates with employers.
They’ve seen a big hit in this since AI, and companies have explicitly cited the fact that AI can complete the same tasks that these junior devs used to perform.
I’m wondering how they really prevent uploads of other peoples faces if they take a clip of a video of another person. I’m sure Apple didn’t open up the 3d Face ID scanning to them to verify
>What's coming next? Bunch of people, with lots of free time watching non-sense AI generated content?
Wasn't this always the outcome of the post labor economy?
For this discussion lets just say that AI+Robots could replace most human labor and thinking. What do people do? Entertainment is going to be the number one time consumer.
It's obvious there is no way OpenAI can keep videos generated by this within their ecosystem. Everything will be fake, nothing real. We are going to have to change the way we interact with video. While it's obviously possible to fake videos today, it takes work by the creator and takes skill. Now it will take no skill so the obvious consequence of this is we can't believe anything we see.
The worst part is we are already seeing bad actors saying 'I didn't say that' or 'I didn't do that, it was a deep fake'. Now you will be able to say anything in real life and use AI for plausible deniability.
It's not that obvious. iOS is pretty secure, if they keep the social network and cameo feature limited to that there might not be good ways to export videos off the platform onto others beyond pointing a camera at the tablet screen. And beyond there being lots of ways to watermark stuff to be detectable, nothing stops the device using its own camera to try and spot if it's being recorded. The bar can be raised quite high as long as you're willing to exclude any device that isn't an iPhone/iPad.
Today's Sora can produce something that resembles reality from a distance, but if you look closely, especially if there's another perspective or the scene is atypical, the flaws are obvious.
Perhaps tomorrow's Sora will overcome the the "final 10%" and maintain undetectable consistency of objects in 2 perspectives. But that would require a spatial awareness and consistency that models still have a lot of trouble with.
It's also possible we remain stuck in the uncanny valley forever, or at least for the rest of our lives.
It's possible to produce some video or image that looks real, cherry-picked for a demo, but not possible to produce any arbitrary one you want that will end up passable.
>Everything will be fake, nothing real. We are going to have to change the way we interact with video.
I'm optimistic here.
Look at 1900s tech like social security number/card, and paper birth certificates. Our world is changing and new systems of verification will be needed.
I see this as either terribly dystopian - or - a possibility for the mass expansion of cryptography and encrypted/signed communication. Ideally in privacy preserving ways because nothing else will make as much sense when it comes to the verification that countries will need to give each other even if they want backdoor registry BS for the common man.
We are going to have to change the way we interact with video.
I doubt it will be for the better. The ubiquity of AI deepfakes just reenforces entrenchment around "If the message reinforces my preconceived notion, I believe it and think anyone who calls it fake is stupid/my enemy/pushing an agenda. If the message contradicts my preconceived notion, it's obviously fake and anyone who believes it is stupid/my enemy/pushing an agenda.". People don't even take the time to think "is this even plausible", much less do the intellectual work to verify.
They can't even be consistent within their own launch video. Consistency is by far the biggest issue with generative AI. How can a professional studio work with scenes which has continuity errors on every single shot? And if it's not targeting professionals, who is it for?
The common thread I am seeing with replacing creative work with AI is that of lowering the bar of acceptability and counterbalancing that with the (potential) savings from taking human labour out of the equation. The cost of labour is not just the raw cost but also the bargaining power that they can exercise by going on strikes etc. From my limited understanding, creatives seem to have more unions than programmers given that I have heard of at least two strikes from voice actors and writers and none from the tech sector. So it should be a win-lose for those who profit off of videos without taking part in the labour process of making one and lose-lose for everyone else.
> counterbalancing that with the (potential) savings
No. It's all about personalization. Even with all the money in the world you couldn't sit a filming crew, VFX specialist, foley artist, and voice actors next to every user of your app, ready to produce new content in 60 seconds.
I don't get why this keeps being framed as a labor thing, it's unlocking genuinely new forms of interactive media.
You write a sentence and get out media... what more personalization are you looking for?
And no, labor is not a factor in the way you tried to frame it.
There is absolutely no one tying up $250,000 in GPUs to let users spit out a funny clip of Sam Altman jumping over a chair because they think that's a smart way a way to get out of paying artists.
>I don't get why this keeps being framed as a labor thing, it's unlocking genuinely new forms of interactive media.
Because it directly impacts people's ability to earn a living. If you truly don't understand this, I think you should spend some time talking to people who are impacted by it. Artists, and so on. Seriously, this is a head-in-the-sand take.
It's very bizarre to act this is being done in order to replace artists.
I build gen AI for entertainment: I don't build to replace anyone, and if my product gets eyeballs existing creators can't, it's because it gives the consumer something they wanted to see in the world.
Past that you're just complaining that consumers don't want what you made.
Myopic view. Just because your motives may be pure and your greed may be a non-issue doesn't mean that multinational corporations who have aligned themselves with fascist regimes (Meta, Alibiba, etc -- point is that they're demonstrably amoral/immoral) do not intend to exploit the tech to its fullest extent.
However having said that, the intention/aim need not be to deliberately replace creatives. That is not the claim I am making or that anyone in this thread or general public discourse is claiming. The minimal claim is simply that the commodification of art decreases the value and employability of people who perform the same task as the AI. It is also not limited to artists. It is all-encompassing. If an employer can now use AI instead of a copywriter, they will often do that -- big and small business alike. The same can be said for many niche fields which previously required specialized education or training.
I am not saying this from an Anti-AI perspective. I own an AI startup.
But use your big boy brain. It is clear that the commodification of intelligence has a downward market pressure on the market value (which is the PERCEIVED value of an employee in the eyes of the EMPLOYER) for many, if not eventually ALL jobs/roles.
I like how you start with "myopic view" the tell me to use my "big boy brain" to see the ease of creation across the board going down as a negative thing.
The purpose of the society was never to pay money to people, and if we figure out how to get grains of sand to replace skilled labor, there's no amount of greed that can outpace what it will do for humanity.
OpenAI apparently assumes that the primary users of Sora 2/the Sora app will be Gen Z, especially with the demo examples shown in the livestream. If they are trying to pull users from TikTok with this, it won't work: there's some nuance to Gen Z interests than being quirky and random, and if they did indeed pull users from TikTok then ByteDance could easily include their own image/video generators.
Sora 2 itself as a video model doesn't seem better than Veo 3/Kling 2.5/Wan 2.2, and the primary touted feature of having a consistent character can be sufficiently emulated in those models with an input image.
Good point. I think OpenAI lacks the cultural understanding that tiktok is providing their users not only with entertainment but also social things like trends, reviews, gossip, self-expression. These aspects are not included in the sora experience.
This is going to sound crass but idc - OAI is just full of geeks, when what is needed is people who are more akin to hippies - thats pretty much what Apple was in the early days.
Its no use building technology when its not married with the humanities and liberal arts.
IMO you're making a valid point, because there seems to be a disconnect between AI and tangible human benefits. The ChatGPT-as-therapy train has been nerfed after the bad publicity, and it is force-fed to people at their workplaces through copilot.
I assume if you ask normal people how AI affects their lifes they'd think about annoying callcenter menus, deep fake porn and propaganda videos, and getting homework done. Not sure if any of this is a positive experience for the mind.
It's 2025 and most speech controls for car navigation don't work, Siri is a pile of sh*t and millionaires are trying to convince us that we should either use their AI or a google which has significantly reduced the quality of their search result pages.
It's like a false choice dilemma which allows back-to-the-roots companies such as Kagi to emerge, and I'm happy about it.
Sheeeeeeeeeeesh. That was so impressive. I had to go back to the start and confirm it said "Everything you're about to see is Sora 2" when I saw Sam do that intro. I thought there was a prologue that was native film before getting to the generated content.
I'm sorry but that's a gross exageration. If any of this was real film then I'd start a gofundme page for OpenAI to get better video production equipment and team because that would be laughably bad.
If anything, it looks a lot worse than a lot of AI-generated videos I've seen in the past, despite being a tech demo with carefully curated shots. Veo 3 just blows this out of the water for example.
It's not an exaggeration to me? I literally stopped the video and went back to the start and re-read. You're more than welcome to speak about your opinions and experiences, but I'm speaking about mine.
I'm over here thinking, "It felt like just yesterday I was laughing at trippy, incoherent videos of Will Smith eating spaghetti."
I love the progress we're making. I love the competition between big companies trying to make the most appealing product demos. I love not knowing what the tech world is going to look like in six months. I love not thinking, "Man. The Internet was a cool invention to have grown up in, but now all tech is mundane and extractive." Every time I see AI progress I'm filled with childlike wonder that I thought was gone for good.
I don't know if this represent SOTA for video generation. I don't care. In that moment I found it impressive and was commenting specifically on the joy I experienced watching the video. I find it frustrating to have that joy met with such negativity.
Don't worry. AI is going to be monetized and extractive in no time. Just like Social Media went from "fresh, fun and cool new tech" to "how did we let this horrible beast take hold of the world," AI will take the same path. In 10 years or sooner, when 99.99% of what you read, hear, and watch is AI slop, you're going to post "This used to be a cool invention!" if there's even a place left for humans to post by that time.
I agree. It will absolutely get there. Such is the trend of all scientific inventions. A breakthroughs occurs, prosperity follows in response, hedonic adaption causes satisfaction to regress to the mean, and then people squeeze every remaining drop of value out of the technology while we wait for those capable of true innovation to work their magic once more. I don't find it idyllic, but I accept it as the way the world works. It feels like a force of nature to me.
The period we're in is fleeting. I think it should be acknowledged and treasured for what it is rather than viewed with disdain because of what is inevitably to come. I stopped using Facebook and never moved to Insta/TikTok when things began to feel too extractive, but, for a good decade there, I felt so close to so many more people than I ever thought possible. It was a really nice experience that I no longer get to have. I'm not mad at social media. I'm happy I got to experience that window of time.
Right now I'm very happy to be using LLMs without feeling like I'm being preyed upon. I love that programming feels fresh and new to me after 15 years. I'm looking forward to having my ability to self-express magnified ten-fold by leveraging generative audio/visuals, and I look forward to future breakthroughs that occur once all these inventions become glorified ad-delivery mechanisms.
None of this seems bad to me. Innovation and technological progress is responsible for every creature comfort I have experienced in my entire life. People deserve to make livings off of those things even if they weren't solely responsible for the innovation.
I fully understand the hype but the initial scene with Sam feels nothing like how any self respecting video producer would create. The jump cuts mid-sentence are extremely jarring, certainly not framed in any traditional sense, and he's almost entirely out of focus.
Points though for the completely expressionless line delivery, it completely nailed that.
You obviously have never actually tried to make anything in AI video.
It is a parlor trick.Maybe this is a big advance but the current state of AI video is a joke.
It is only impressive if you don't actually make anything. It is impressive in a marketing release that is quickly forgot about.
Will Smith eating spaghetti is the dumbest most uncreative thing. You are impressed by it because it is a meme. It is stupid.
I didn't claim to have made anything with AI video. I'm just commenting on how rapidly things appear to be improving from an external viewpoint. We used to give AI crap for failing to generate an appropriate number of fingers on still imagery. Now we're watching multiple minutes of video to find handfuls of discontinuities. The goalposts have shifted pretty far in an exceptionally short amount of time.
I have no idea why you're so intent on coming across bitter about a fledgling technology. A few years ago this demo video would've been indistinguishable from magic. It will continue to improve.
I get what you're saying. I see it too. That said, a lot of people won't notice the flaws, especially with these fast, choppy cuts. By the time you realize the neck is way too long or whatever, its 2 cuts later.
Anecdotally, I forgot multiple times that I was watching AI generated content, and my partner tuning in and out of it asked me a few times if we were watching the demo (as opposed to the real video). We are both pretty sensitive to slop too. I think something has flipped here.
Let me tell you, I wish I was paid by OpenAI, but unfortunately I get paid a very average wage in a country where OpenAI do not operate.
You can look at my account history and see that there is absolutely no indication that I am paid for, or a bot, I just happen to be relatively enthusiastic and bullish on AI capabilities. My account is 5 years old.
I do not think this product from OpenAI will be good for the world, I think it is mostly a very bad thing. I just think it is impressive.
I encourage you to try make the most generous interpretation of my comment and try to consider that other people can have sometimes wildly different experiences to you, however hard that is to believe. The alternative is to accuse people of being paid for commentors, which benefits no one and just isolates you further.
To add more nuance to my original comment, yes if you look closely at the videos it is clear that they are AI generated, but as I was watching I 'lapsed' out of attention a few times and had to remind myself what I was looking at was generated. If I stopped and rewound a few times, yes of course I could notice the slop. But my point is that it's the first time I have lost the "I am watching slop" thought while watching AI video.
Hope that clears things up for you.
Edit: I just looked through your comment history, you are an extreme AI perma bear, and that's ok, that doesn't mean you're a paid for anti-ai lobbyist. Also every single comment you have made is extremely negative and inflammatory.
I encourage you to revise your communication methods and try to practice some kindness.
No, the average person just doesn't have a creative bone in their body.
They will be impressed when they get Super Hero movie #1899 but this time by AI , woo hoo!
I am creative and have released music with a large listenership. The other person in my original comment is a designer by trade.
Anything else dismissive you’d like to say or do you enjoy leaving snarky comments?
Anyone with access able to confirm if you can start this with a still image and a prompt?
The recent Google Veo 3 paper "Video models are zero-shot learners and reasoners" made a fascinating argument for video generation models as multi-purpose computer vision tools in the same way that LLMs are multi-purpose NLP tools. https://video-zero-shot.github.io/
It includes a bunch of interesting prompting examples in the appendix, it would be interesting to see how those work against Sora 2.
One use that occurred to me is that fans will be able to "fix" some movies that dropped the ball.
For example, I saw a lot of people criticizing "Wish" (2023, Disney) for being a good movie in the first half, and totally dropping the ball in the last half. I haven't seen it yet, but I'm wondering if fans will be able to evolve the source material in the future to get the best possible version of it.
Maybe we will even get a good closure for Lost (2004)!
(I'm ignoring copyright aspects, of course, because those are too boring :D)
> Maybe we will even get a good closure for Lost (2004)!
Whether it's text or super-advanced VR holograms, if it's fan fiction it's fan fiction. Which can be interesting and compelling, but that will never be as exciting as the Word of God[0]. Death of the Author is a nice thought experiment but few people really adhere to it, I've found.
Or just going to the Goofs section of a movie on IMDB, and fix the trivial issues (e.g. car had cracked window in earlier scene, and suddenly a normal window in another scene).
My issue is that the copyright aspect are what prevents me from using this as much as I otherwise would.
About 6 months ago I asked a few different AIs if they could translate a song for me as a learning experience, meaning not a simple translation, but more a word by word explanation of what each word meant, how it was conjugated, any more musical/lyrical only uses that aren't common outside of songs, and so on. I was consistently refused on copyright grounds, despite this seeming a fair use given the educational nature. If I pasted a line of the lyrics at a time, it would work initially, but eventually I would need to start a new chat because the AI determined I translated too much at once.
So in this one, if I wanted to ask it to create a video of the moment in Final Fantasy 6 when the bad guy wins, or a video of the main characters of Final Fantasy 7 and 8 having a sword duel, would it outright refuse for copyright reasons?
It sounds like it would block me, which makes me lose a bit of interest in the technology. I could try to get around it, but at what point might that lead to my account being flagged as a trouble maker trying to bypass 'safety' features. I'm hoping in a few years the copyright fights on AI dies down and we get more fair use allowance instead of the tighter limitations to try to prevent calls for tighter regulation.
> You can sign up in-app for a push notification when access opens for your account.
You need to be in the US/Canada and wait for this notification, and when you get an invite you can start using it in the app and on sora.com. And apparently you get 4 more invite codes that you can share with anyone, e.g. Android users:
> Android users will be able to access Sora 2 via http://sora.com once you have an invite code from someone who already has access
It's wild that I have a paid account but I have to scour the Internet to find someone else with a paid account and beg them for an invite code to use the product I already paid for. Make it make sense.
Sora 1 was also lauded as being incredibly good at physics based on the early cherry-picked examples. The phrase "world simulator" was thrown around a lot. That didn't last long once people finally got their hands on it though.
Kind of wondersome if they will start to combine LLM generation with actual world models/GPU engines. Imagine that your model generates the wireframes, the Engine generates the physics and then another model fills in the actual visuals, and gaps... So you have realistic physics and gaps are filled in... Will also help with image retention more, if objects moved behind each other.
It was so much more hyped than that. They made it sound like Hollywood was in big trouble.
It is going to have the same problems as Midjourney. You just don't have that much control of the scene. The process is to make thousands of random variations and cherry pick the good stuff because you can't do anything else.
The demo on their homepage shows really bad physics. There’s a lot of it, but that doesn’t mean it’s correct. The hair of Sam looks like a paper cutout in almost every shot.
I received an invite and spent the last 6 hours generating an ungodly amount of free videos on the Sora app, they are AS GOOD as the cherry picked ones in the demo. They are insane. And the "social aspect" of the app is also very good.
I predict that this will move some people over, and IG/TT will lose marketshare.
It's obvious there is no way OpenAI can keep videos generated by this within their ecosystem. Everything will be fake, nothing real. We are going to have to change the way we interact with video. While it's obviously possible to fake videos today, it takes work by the creator and takes skill. Now it will take no skill so the obvious consequence of this is we can't believe anything we see.
The worst part is we are already seeing bad actors saying 'I didn't say that' or 'I didn't do that, it was a deep fake'. Now you will be able to say anything in real life and use AI for plausible deniability.
I predict a re-resurgence in life performances. Live music and live theater. People are going to get tired of video content when everything is fake.
In light of some comments and videos here, I’d like to morbidly announce that I can no longer distinguish between AI videos and real ones. However, I’ll take this as an opportunity to move from short-form content to long-form, since it seems that space hasn’t yet been hijacked by AI.
That could totally power next generation of green-screen techs. Generative actors may not find favorable response in the audiences; but SFX, decor, extras, environments that react to actors' actions - amazing potential.
There's less here than you think. Video games have already been procedurally generating environment art for quite some time, and film/tv are already leveraging that with giant screens that use Unreal Engine to create the backgrounds.
AI could be helpful here, but it's not clear that it is required or an improvement.
I don't get the autotune argument. It's like saying we shouldn't be using electronic instruments because it's not real or we shouldn't use digital audio instruments because they're not real etc.
It's just a way to get different kind of sound. It won't make you good tracks.
I find this one interesting because Rap has classically been difficult for these models (I think because it's technically difficult to find the right rhythms and flow for a given set of lyrics).
I don't know but it doesn't impress me one bit? like I'm not trying to hate, but it just seems kind of like the model is given the track and then it tries to just follow it by matching words and then spitting them out, like as if it could talk about making a sandwich over some epic track and it'd sound the same?
like, LLMs are fantastic at generating patterns, so words that match and same with images etc.
But there's not much uniqueness? it's "impressive" like a savantic kind of ability to come up with rap, but it doesn't really product something I'd want to listen to..?
I listened to the metal thing and kind of the same thing?
It's very high fidelity, like the quality of the drums and etc it's quite impressive, but the vocals seem off? it's like a poem being read by TTS then transformed into "metal voice"
and kind of just an averaging of "metal music" kind of like stock photos and into a track, very formulaic
not to mention many metal bands etc they do formulaic stuff especially if they have an identifying kind of hit
But to me this is cool tech, but I wouldn't listen to it
I've listened music for a long time but I don't listen to a wide variety today, however for example with pop it can be very complex or very simple, but average or "almost" will really not make a good song, it can seem simple in hindsight but probably blood sweat and tears went into such songs, or creative energy that might never come back as strong.
just my raw thoughts though. it could be me being biased knowing it's AI, but I don't think so. I think my brain has kind of adapted to a point where I can feel if something is AI because it always seems super "average"/mid?
I'd love to see a blind study comparing a wide spectrum of these AI tracks to lesser known real artists (so the participants don't just recognize the songs) to see whether people can tell or if knowledge of the source biases them. I'm genuinely curious as to the results.
I don't think people would think anything strange of a lot of these tracks if they just randomly heard them on the radio.
When you listen to music that has been AutoTuned, you don't know if the singer can actually... sing. If you put them in a room and asked them to sing a song without artificial aid, would you actually enjoy their performance or not? You don't know!
This marked a divergence from thousands of years of vocal performances where singing ability and enjoyment of the music were one and the same.
AutoTune was the first slop, and the general population seems to like it.
The arguments against auto-tune are typically different, since it's obvious to anyone that autotune can't make you sound like a soprano if you're nowhere near - so skill is still required.
The problem with autotune is more that it removes a lot of nuance from singers' voices, it's like listening to MIDI instead of listening to a real piano. This is, however, something that can be improved. Synthesizers can produce wonderful musical effects, and there's lots of highly virtuoso music on synthesizers (including voice distortions, pretty similar technically to autotune) for those that are into it. Progrock, for example, was all about using new technology in complex and extremely interesting ways. Maybe more interestingly for your particular objection, you can look at early electronic music, say Vangelis or Isao Tomita or Kraftwerk. For at least parts of their songs, they could have just programmed their synthesizers ahead of time and played concerts without even being on stage - but that doesn't take away from the fact the music itself.
Ultimately, if the music sounds good and elicits some feelings and thoughts, it's good music. Whether the musicians can reproduce it live or it's done 90% in a studio doesn't really matter here. Of course, it does mean it may not be worth going to a live show from some particular performer, and it also means that the performer is not necessarily the most relevant artist - the person programming the "auto"tune should at least be considered part of the band.
That's like saying movies are not good cause they're not live action-only performances
For me the biggest thing is actually the production, there's many people involved usually and sometimes real magic gets made, and that magic might not even contain any vocals at first
like what is acceptable music? only raw vocals & acoustic instruments?
At least in terms of realism, the image generation field is at the realism line now. Single frame generation with Wan 2.1 / 2.2 (and others) for example, will get you realism.
Not related to Sora but, I have been looking for / hoping for an AI powered motion tracking solver. I've used Blender and Mocha in AE and both still require quite a bit of manual intervention, even in very simple scenes.
I saw some promnise with the Segment Anything model but I haven't seen anyone yet turn it into a motion solver. In fact I'm not sure if can do that at all. It may be that we need to use an AI algorithm to translate the video into a more simple rendition (colored dots representing the original motion) that can then be tracked more traditionally.
Sam Altman has made (for me) encouraging statements in the past about short-form video like TikTok being the best current example of misaligned AI. While this release references policies to combat "Doomscrolling and RL-sloptimization", it's curious that OpenAI would devote resources to building a social app based on AI generated short form video, which seems to be a core problem in our world. IMO you can't tweak the TikTok/YouTube shorts format and make it a societal good all of a sudden, especially with exclusively AI content. This is a disturbing development for Altman's leadership, and sort of explains what happened in 2023 when they tried to remove him... -> says one thing, does the opposite.
>IMO you can't tweak the TikTok/YouTube shorts format and make it a societal good all of a sudden, especially with exclusively AI content.
I agree. At best, short videos can be entertainment that destroys your attention span. Anything more is impossible. Even if there were no bad actors producing the content, you can't condense valuable information into this format.
I'm optimistic about the Sora app! My hope is that it becomes much more whimsical and fun than TikTok because everyone on the app knows that all content is fake. Hopefully that means less rage-bait and more creative content, like OG YouTube. Nobody's going to get their news from Sora because it's literally 100% fake.
In the early years everyone told me that TikTok is actually fun and whimsical (like just after it stopped being musical.ly), and it's all about fun collaboration, and amateur comedy sketches, fun dances and lipsyncs, and people posting fun reactions to each other etc, all lighthearted and that social media is finally fun again!
Why would it be more like OG YouTube, when the content they demoed very closely resembles YouTube shorts? The key difference is OG YouTube was long form.
Sam Altman is a businessman. His job is to say whatever assuages his market, and that includes gaslighting you when you're disgusted by AI.
If you never expected Altman to be the figurehead of principled philosophy, none of this should surprise you. Of course the startup alumni guy is going to project maligned expectations in the hopes of being a multi-trillion dollar company. The shareholders love that shit, Altman is applying the same lessons he learned at Worldcoin to a more successful business.
There was never any question why Altman was removed, in my mind. OpenAI outgrew it's need for grifters, but the grifter hadn't yet outgrown his need for OpenAI.
Nope. A CEO can't essentially steal from shareholders, but otherwise they have extremely broad latitude in how they engage in business.
There is no legal or moral imperative to make antisocial, unethical, or short term decisions that "maximize shareholder value."
This is something that morally weak people tell themselves (and others) to justify the depravity they're willing to sink to in order to satiate their greed.
The concept doesn't even make sense: different shareholders have different priorities and time horizons. A businessperson has no way to know what it objectively means to maximize their returns. They must make a subjective determination, and they have extremely broad latitude to do that.
If I run an AI business, then people using more AI means more business. If noone uses my AI then I go out of business
Increasing shareholder value can be done in the broadest sense by just increasing business
If I fund my own business, I can control growth and _choose_ ethics over profits, in the hope that stunting growth is acceptable if my customers value ethics too, and that whomever I someday pass my company to shares these values
If I take capital investment, I now have a contractual agreement to provide returns on that investment. Yes failure to adhere can result in lawsuits or legal penalties. Or I can be fired/voted out for failing to bring high enough returns. I now _cannot_ choose ethics over profits, due to the conflict of interest of self-preservation
So you are correct - there is no legal or moral contract to behave unethically, but there is instead a strong systemic and self-preserving incentive to do so
I think we almost agree here, but you make it sound as if the exec can simply stand up and do the right thing here. I argue the exec will simply be pushed aside for another
This is what people refer to when they talk about the binds that hold modern day mega-corps
If you yourself are an exec, I personally think you can understand these truths and work with them as best you can, and still be a good human being of course, but that there are lines that should not be crossed to keep a job
It is a collective issue we need to solve that of course starts with each individual seeing the true situation with kindness and compassion
You’re just saying there are incentives for unethical behavior? Yeah, obviously.
They don’t need to be excused by “well that’s their obligation.” It’s not! Actually, a person’s obligation is to act morally even when there are incentives otherwise, which is approximately all the time for nearly every person.
This is something children learn (lest they be excluded from their society) yet Very Smart People in the upper echelons of the business world conveniently forget.
> If I take capital investment, I now have a contractual agreement to provide returns on that investment. Yes failure to adhere can result in lawsuits or legal penalties.
This is not true. If you've signed a contract that says anything like this, consider getting a real lawyer.
Someone who doesn't follow the moving edge would be forgiven for being confused by the dismissive criticism dominating this thread so far.
It's not that I disagree with the criticism; it's rather that when you live on the moving edge it's easy to lose track of the fact that things like this are miraculous and I know not a single person who thought we would get results "even" like this, this quickly.
This is a forum frequented by people making a living on the edge—get it. But still, remember to enjoy a little that you are living in a time of miracles. I hope we have leave to enjoy that.
I remember having to tell people "Don't describe too many individual objects in your DALL-E 2 prompt, because the model has trouble separating discrete concepts and the features tend to blend together."
Now we have photorealistic video with sound and, oh yeah, the model can generate an entire script and mini-plot on its own based on the most basic prompt.
I know the comments here are gonna be negative but I just find this so sick and awesome. Feels like it's finally close to the potential we knew was possible a few years ago. Feels like a pixar moment when CG tech showed a new realm of what was possible with toy story
These videos are a very impressive engineering feat. There are a lot of uses for this capability that will be beneficial to society, and in the coming years people will come up with more good uses nobody today has thought of yet.
But clearly we also see some major downsides. We already have an epidemic of social media rotting people's minds, and everything about this capability is set to supercharge these trends. OpenAI addresses some of these concerns, but there's absolutely no reason to think that OpenAI will do anything other than what they perceive as whatever makes them the most money.
An analogy would be a company coming up with a way to synthesize and distribute infinite high-fructose corn syrup. There are positive aspects to cheaply making sweet tasting food, but we can also expect some very adverse effects on nutritional health. Sora looks like the equivalent for the mind.
There's an optimistic take on this fantastic new technology making the world a better place for all of us in the long run, after society and culture have adapted to it. It's going to be a bumpy ride before we get there.
I actually wonder if this will kill off the social apps and the bragging that happens. It will be flooded by people faking themselves doing the unimaginable.
This is also my thesis. The internet is going to be saturated with AI slop indiscernible from real content. Once it reaches a tipping point, there will no longer be much of a reason to consume the content at all. I think social networks that can authenticate video/photo/text content as human-created will be a major trend in a few years.
I have no clue if the reactions are real, but there are some videos online of people showing their grandparents gameplay from Grand Theft Auto games trying to convince them that it is real footage. The point of the videos is to laugh at their reactions where they question if it really happened, etc.
Maybe this will result in something similar, but it can affect more people who aren’t as wary.
I regularly get AI movie recaps on my shorts and I just eat it up.
The very fact that I (or billions of others) waste time on shorts is an issue. I don't even play games anymore, it's just shorts. That is a concerning rewiring of the brain :/
Guess what I`m trying to say is that, there is a market out there. It's not pretty, but there certainly is.
Will keep trying to not watch these damn shorts...
Yes, I wonder if the content distribution networks that call themselves "social networks" can even survive something like this.
Of course, the ones focusing on the content can always editorialize the spam out. And in real social networks you ask your friends to stop making that much slop. But this can be finally the end of Facebook-like stuff.
> There are a lot of uses for this capability that will be beneficial to society
Are there? “A lot” of them? Please name a few that will be more beneficial than the very obvious detrimental uses like “making up life-destroying lies about your political opponents or groups of people you want to vilify” or “getting away with wrongdoing by convincing the judge a real video of yourself is a deepfake”.
It can generate funny videos of bald JD Vance and Harry Potter characters for TikTok. Which makes me wonder, what is the actual plan to make money off these models? Billions have been invested but the only thing they seem to be capable of is shitposting and manipulation. Where is the money going to come from?
> There are a lot of uses for this capability that will be beneficial to society
Please enlighten me. What are they? If my elderly grandma is on her deathbed and I have no way to get to see her before she passes, will she get more warmth and fond memories of me with a clip of my figure riding an AI generated dragon saying goodbye, or a handwritten letter?
Still zero responses, eh? My example was charged but I clearly had a point: how does AI fill a void where meaning should be, over what has worked for centuries? How is it better than face to face, or a handwritten letter?
Suno allows me to rapidly flesh out demos and brainstorm. Played music my whole life manually. Easier for me to find what I'm looking for and while avoiding demo love.
I still feel this is limited by what it learned from. It looks cool but it also looks like something I'd dreamt or saw flicking through TV channels. Kind of like spam for the eyes.
No doubt they can create Hollywood quality clips if the tools are good enough to keep objects consistent, example, coming back to the same scene with same decor and also emotional consistency in actors
I think this is not nearly as important as most people think it is.
In hollywood movies, everyone already knows about "continuity errors" - like when the water level of a glass goes up over time due to shots being spliced together. Sometimes shots with continuity errors are explicitly chosen by the editor because it had the most emotional resonance for the scene.
These types of things rarely affect our human subjective enjoyment of a video.
In terms of physics errors - current human CGI has physics errors. People just accept it and move on.
We know that superman can't lift an airplane because all of that weight on a single point of the fuselage doesn't hold, but like whatever.
Location consistency is important. Even something as simple and subtle as breaking the 180-rule [1] feels super uncanny to most audiences. Let alone changing the set the actor occupies, their wardrobe, props, etc.
There are lots of tools being built to address this, but they're still immature.
Well put. Honestly the actor part is mostly solved by now, the tricky part is depicting any kind of believable, persistent space across different shots. Based off of amateur outputs from places like https://www.reddit.com/r/aivideo/, at least!
This release is clearly capable of generating mind-blowingly realistic short clips, but I don't see any evidence that longer, multi-shot videos can be automated yet. With a professional's time and existing editing techniques, however...
I wonder if this stuff is trained on enough Hallmark movies that even AI actors will buy a hot coffee at a cafe and then proceed to flail the empty cup around like the humans do. Really takes me out of the scene every time - they can't even put water in the cup!?
No way man, this is why i loved Mr Robot, they actually payed a real expert and worked story around realism and not just made up gobbleygook that shuts my brain off entirely to its nonsense
Cool demo! But let’s pour one out for all the weird, janky, hand crafted videos that made early internet so fun. Anyone else still crave that kind of content?
The ability for the masses to create any video just by typing, among the other features, is not novel technology? Or is it just the lack of emotional response?
> How the FUCK does Sora 2 have such a perfect memory of this Cyberpunk side mission that it knows the map location, biome/terrain, vehicle design, voices, and even the name of the gang you're fighting for, all without being prompted for any of those specifics??
> Sora basically got two details wrong, which is that the Basilisk tank doesn't have wheels (it hovers) and Panam is inside the tank rather than on the turret. I suppose there's a fair amount of video tutorials for this mission scattered around the internet, but still––it's a SIDE mission!
Everyone already assumed that Sora was trained on YouTube, but "generate gameplay of Cyberpunk 2077 with the Basilisk Tank and Panam" would have generated incoherent slop in most other image/video models, not verbatim gameplay footage that is consistent.
> Everyone already assumed that Sora was trained on YouTube
Doesn't this already answer your question...? "Let's Play" type videos and streams have been a thing for years now, even for more obscure games. It very well could've been trained on Cyberpunk videos of that mission.
It's hard for me to believe that the model coherently memorized both the video and audio of a relatively obscure Let's Play, and that a simple prompt was enough to surface it (the use of the term "Basilisk tank" would also likely not be in video metadata either). That is the reason the person who made that tweet, who has far more prompting experience than myself, was shocked.
That's really interesting. What if they RAG search related videos from the prompt, and condition on that to generate? That might explain fidelity like this
An interesting counterexample is "a screen recording of the boot screen and menus for a user playing Mario Kart 64 on the N64, they play a grand prix and start to race" where the UI flow matches the real Mario Kart 64, but the UI itself is wrong: https://x.com/fofrAI/status/1973151142097154426
Their anime vid gen is really, really impressive. The results I've seen aren't /good/ from an industry-standard (nothing compared to the likes of the Demon Slayer movie I watched in theatres recently), but I legitimately couldn't tell that it was AI-generated. Massive step up from Sora 1 and other vid gen models.
Here's to hoping that the industry will adapt to have it aid animators for in-betweening and other things that supplement production. Anime studios are infamously terrible with overworking their employees, so I legitimately see benefits coming from this tool if devs can get it to function as proper frame interpolation (where animators do the keyframes themselves and the model in-betweens).
I'm a software engineer and hobbyist actor/director. My friends are in the film industry and are in IATSE and SAG-AFTRA. I've made photons-on-glass films for decades, and I frequently film stuff with my friends for festivals.
I love this AI video technology.
Here are some of the films my friends and I have been making with AI. These are not "prompted", but instead use a lot of hand animation, rotoscoping, and human voice acting in addition to AI assistance:
I see several incredibly good things happening with this tech:
- More people being able to visually articulate themselves, including "lay" people who typically do not use editing software.
- Creative talent at the bottom rungs being able to reach high with their ambition and pitch grand ideas. With enough effort, they don't even need studio capital anymore. (Think about the tens of thousands of students that go to film school that never get to direct their dream film. That was a lot of us!)
- Smaller studios can start to compete with big studios. A ten person studio in France can now make a well-crafted animation that has more heart and soul than recent by-the-formula Pixar films. It's going to start looking like indie games. Silksong and Undertale and Stardew Valley, but for movies, shows, and shorts. Makoto Shinkai did this once by himself with "Voices of a Distant Star", but it hasn't been oft repeated. Now that is becoming possible.
You can't just "prompt" this stuff. It takes work. (Each of the shorts above took days of effort - something you probably wouldn't know unless you're in the trenches trying to use the tech!)
For people that know how to do a little VFX and editing, and that know the basic rules of storytelling, these tools are remarkable assets that compliment an existing skill set. But every shot, every location, every scene is still work. And you have to weave that all into a compelling story with good hooks and visuals. It's multi-layered and complex. Not unlike code.
And another code analogy: think of these models like Claude Code for the creative. An exoskeleton, but not the core driving engineer or vision that draws it all together. You can't prompt a code base, and similarly, you can't prompt a movie. At least not anytime soon.
Taking the time and effort out of something is exactly what strips it of its beauty
Beauty is not just an “idea” that someone has and needs to get out onto a medium
It is a process and journey that a person undergoes to get said idea onto said medium
That journey often plays out very differently than the person expects. Things change, the art is different from the idea, and the person learns and grows
Our modern society is so obsessed with results, competition, and efficiency that we no longer see the truth: the journey is to be enjoyed, and from enjoying the journey, comes beauty
I encourage you to meditate on why our society is so sick and depressed right now, and extrapolate to how we got here, before assuming this will be a good thing for society
I saw a quote earlier this week that I'll copy here:
> I considered renting out sound stages, flying to exotic desert locations, getting a scuba team to shoot the underwater scenes in an aquarium, commissioning custom-made Teletubbies costumes, hiring SAG actors, building dozens of miniature sets, and spending my life savings on making this video. But using AI just seems slightly easier.
Making short films with AI is still incredibly effortful. If you're being careful and diligent, it takes days to "shoot" and edit the entire shot list for a 5-7 minute short.
Would you say that the creators of today's animated TV shows, in mechanizing production with Toon Boom Studio, have stripped the beauty away? I still found "Bojack Horseman" to be a salient dramedy.
Would you say that Pixar, in using motion capture and algorithms to simulate light, physics, and movement, is cutting away the journey?
This is a new adventure and new level of abstraction we're embarking upon.
I'm already thinking about the next way points: real time mocapped improv for D&D campaigns and live community theater fantasy and science fiction productions.
These are tools that bring us to new places, that enable us to tell new stories. Previously you'd have to win Disney budget approval to tell a story matching your vision - now you don't.
But I will still be entertained. Expedient AI expression can touch most people the same way a low effort meme or an off the cuff whitticism.
Art is not effort. Art is not labour. Beauty is not suffering. Art =/= craft. Art is communication.
If someone wants to suffer long the endurance journey to becaome skilled at a craft we can still respect/appreciate it the same way a sprinter spends 10 years training to run real fast, in the mean time most of us will use a vehicle to get somewhere faster.
What we're going to lose is a bunch of interesting behind the scene videos because no one is going to watch someone prompt for an hour wondering why can't I do that, but rather why didn't I do that.
Proliferating tools for creation is net good in the same sense that teaching masses to write is net good. It's strange people are opposing lowering the barrier to entry to visual communication. That's what art ultimately is, communication. Once difficult, soon ubiquituous.
Creative people with ambition and limited resources make good things today without this technology. All this does is accelerate the rate at which low quality "content" is produced by people that have no interest in learning a craft, without attribution and without compensation for the people that have made the effort and whose works train these models.
We mustn't teach peasants how to read or write or else one day we'll live in hellscape of infinte unread inboxes and eternal september... both of which sucks, but much less than a world where masses were illiterate. Ai art/slop is just that for visual communication. Now it's supremely shitty this power is currently being monetized / controlled by a few, the same way communication material like papyrus and paper was jealously guarded/exploited before printing press proliferated, then the mistake that was the internet giving desemination power to every pleb. But there's free models out there and maybe (hopefully) in the not too distant from now the barrier to entry is accessible commodity level hardware and artists just have to eat shit and realize they've contributing to the creative common/canon like those before, i.e. world where monks/literati simply copied and duplicated work in pre copyright era because knowledge/expression, unless jealously guarded, was a collective resource to be built upon.
"The studios and creators who thrive in this new landscape will be those who can effectively harness AI’s capabilities while maintaining the human creativity and vision that ultimately drives the art of cinema."
It is in many ways thrilling to see this come to life, and I couldn't agree with you more.
> "The studios and creators who thrive in this new landscape will be those who can effectively harness AI’s capabilities while maintaining the human creativity and vision that ultimately drives the art of cinema."
..Just somehow several years on, these optimistic statements still all end up being in the future tense, somehow for all the supposed greatness and benefits, we still dont see really valuable outputs.
A lot of us do not want more of the "CONTENT" as envisioned by corporate ghouls who want their employees or artists to "thrive" (another word kidnapped by LinkedIn-Linguists). The point is not in the speed and easiness of generation of outputs, visual and sound effects etc. The point is the artists interpretation and their own vision, impressions etc. Not a statistical slop which "likely" fits my preferences (i.e. increases my dopamin levels).
Adding to the list: The Adventures of Reemo Green. Very funny, and the first time I’ve watched AI video and enjoyed it as more than a technical curiosity.
I'll have you know that in this year's Atlanta 48 Hour Film project (something I've been doing since I was a teen), several teams used AI.
Rewind to just one year prior -- 2024.
AI video was brand-spanking new. We'd only just gotten over the "Will Smith" spaghetti video and the meme-y "Pepperoni Hug Spot" and "Harry Potter by Balenciaga" videos.
I was the only person to attempt to use AI in 2024's competition. It was a time when the tools and infrastructure for video barely existed.
On the debut night, I was resoundingly booed by the audience. It felt surreal. Working all weekend to have an audience of peers jeering at you in a dark theater. The judges gave me an award out of sympathy.
Back then, image-to-video models really were not a thing (Luma launched "Dream Machine v1" shortly after this). I was using Comfy, Blender, Mocap, a full Mocap suit (the itchy kind), and a lot of other hacks to build something with extremely crude tools.
We lost a day of filming and had to scramble to get something done in just 24 hours. No sleep, too much caffeine. Lots of sweat and toil.
The resulting film was a total mess, of course:
https://vimeo.com/955680517/05d9fb0c4f (It's seriously bad - I hate it. It might legitimately be the very first time AI was used in a 48 hour competition.)
That said, it felt very much like a real 48 Hour competition to me. Like a game jam. The crude ingredients, an idea, the clock. The hustle. The corners being cut. It was palpable.
I don't think you can say there isn't soul in this process. The process has so much soul.
Anyway, fast forward to this year. Three teams used AI, including my own. (I don't think I have a link to our film, sadly.)
We all got applause. The audience was full of industry folks, students, and hobbyists. They loved it. And they knew we used AI.
The industry is anxious but curious about the tech. But fundamentally, it's a new tool for the tool box. The real task is storytelling.
Carter needs a new laptop. His daily driver has been falling apart for ages but he refuses to give it up.
We all told him about the sound mix - he let a couple of videos slip with a bad "mono as single-channel stereo audio" renders. On his machine it sounded normal. He got flack for that, and he's been hearing this for months.
I'm going to show him this thread. I don't think he'll ever forget to check again.
Despite that, he's a really talented guy. Chalk this up as a bad production deploy. We didn't want to delete and re-upload since the videos had legs when we first released them. There's a checklist now.
That's the first thing I noticed, too. The first words you hear in the trailer sounds like someone ran the voice through a comb filter. It's so bad it made my skin crawl immediately.
It's the inaccuracy of things like shadows, sub-surface scattering and specular highlights. I think the shadow inaccuracy is what the human visual system is most sensitive to.
These LLMs might make content that looks initially impressive but they are absolutely not performing physically based rendering or have any awareness of the lighting arrangement in these scenes. There are a lot of things they get right, but you only have to screw up one small element to throw the whole thing off.
I am willing to bet that Unreal Engine 5 will continue to produce more realistic human faces than OAI ever can with these types of models. You cannot beat the effects of actually running raytracing in a PBR pipeline.
Yeah, the faces aren't right, and impressive as it is I'm getting icky "uncanny valley" vibes from this.
CGI for fantasy stuff is unavoidable, but when it's stuff that could have been done by actors but is instead AI, then to me it just feels cheap and nasty - fake.
Going to be an amazing source of training data, wait till they get it to real time and people are leaving their video camera open for AR features. OpenAI is about to have a lot of current real world image data, never mind the sentiment analysis.
I don't think they were limited for video training data. Gathering real world data is pretty easy, gathering curated information is a little more difficult.
IDK if the site is being hugged to death but I can only load the first video. Even in just one viewing there were noticeable artifacts, so my impression is that Veo is still in the lead here.
> A lot of problems with other apps stem from the monetization model incentivizing decisions that are at odds with user wellbeing. Transparently, our only current plan is to eventually give users the option to pay some amount to generate an extra video if there’s too much demand relative to available compute. As the app evolves, we will openly communicate any changes in our approach here, while continuing to keep user wellbeing as our main goal.
> And we're introducing Cameo, giving you the power to step into any world or scene, and letting your friends cast you in theirs.
How much are they (and providers of similar tools) going to be able to keep anyone from putting anyone else in a video, shown doing and saying whatever the tool user wants?
Will some only protect politicians and celebrities? Will the less-famous/less-powerful of us be harassed, defamed, exploited, scammed, etc.?
it seems like this is basically youtube's ContentID, but for your face. as long as you upload your "cameo" aka facial scan to them, they can recognize and control the generation of videos with it. if you don't give them your face, then they can't/won't.
"Consent-based likeness. Our goal is to place you in control of your likeness end-to-end with Sora. We have guardrails intended to ensure that your audio and image likeness are used with your consent, via cameos. Only you decide who can use your cameo, and you can revoke access at any time. We also take measures to block depictions of public figures (except those using the cameos feature, of course). Videos that include your cameo—including drafts created by other users—are always visible to you. This lets you easily review and delete (and, if needed, report) any videos featuring your cameo. We also apply extra safety guardrails to any video with a cameo, and you can even set preferences for how your cameo behaves—for example, requesting that it always wears a fedora."
Until you have 2 people that are near identical. They don’t even have to be twins, there are plenty of examples where people can’t even tell other people apart. How is an AI going to do it?
You don’t own your likeness. It’s not intellectual property. It’s a constantly changing representation of a biological being. It can’t even be absolutely defined— it’s always subject to the way in which it was captured. Does a person own their likeness for all time? Or only their current likeness? What about more abstract representations of their likeness?
The can of worms OpenAI is opening by going down this path is wild. We’re not current able to solve such a complex issue. We can’t even distinguish robots from humans on the internet.
If this company's guardrails end up sufficiently working well in practice (note phrases like "intended", "take measures", and "preferences...requested", on things they can't do 100%)... there will be weak links elsewhere, letting similar computation be performed without sufficiently effective guardrails against abuse?
How do we prepare for this? Societal adjustment only (e.g., disbelieving defamatory video, accepting what pervs will do)? Establishing a common base of cultural expectations for conduct? Increasing deterrence for abusers?
Looks like it requires you to film yourself from specific angles and while repeating an autogenerated phrase. Like a pre-AI selfie taken with a handwritten placard with your username and the date or whatever.
Normalizing effective abolishment of consent for imagery, or just consent in general for just about anything, when it can portray anyone doing anything.
If Deepfakes remain the tools of nation state actors, laypeople will be easily fooled.
If Deepfakes are available on your iPhone and within TikTok, everyone will just ask "Is it Photoshop?" for every shred of doubt. (In fact, I already see people saying, "This looks like AI".)
This is good. Normalize the magic until it isn't magic anymore.
People will get it. They're smart. They just need exposure.
> People will get it. They're smart. They just need exposure.
I really doubt this.
If you are in the creative field, your work will just be reduced to "is this slop?" or "fixed it!" with a low effort AI generated work of your original work (fuck copyright right?).
I already see artists battling and fighting putting out their best non AI work only for their audience to question if it is real and they lose the impressiveness.
This just already undermines creators who don't use AI generated stuff.
But who cares about them right? "it is the future" and it is most definitely AGI for them.
But then again, the starving artist never really made any money and this ensures that the artform stays dead.
> People will get it. They're smart. They just need exposure.
It's either this, or the opposite (eg, misinformation needs to be censored). Seems like we as a society can't quite make up our mind on which approach to take.
Clicking a link on the OpenAI dashboard and beeing greeted with a full page of scandily clad women was certainly not what I expected to see when opening Sora..
Ok that's technically really impressive, and probably totally unusable in a real creativity context beyond stupid ads and politically-motivated deepfakes.
I love the casual reminds that we're second-class citizens each time a new technology gets released. Available in the US but always excluding Puerto Rico.
The result of that isn't rational skepticism, though. It's distrusting mainstream news and embracing whatever conspiracy theories your friends believe.
Doing this as a social app somehow feels really gross, and I can't quite put to words why.
Like, it should be preferable to keep all the slop in the same trough. But it's like they can't come up with even one legitimate use case, and so the best product they can build around the technology is to try to create an addictive loop of consuming nothing but auto-generated "empty-calories" content.
I see it more as recognizing that it will take time to be good enough for other use cases so for this release they are targeting it as just something to have fun with. After seeing LLMs crammed into everything whether it makes sense or not, I can appreciate that.
What makes TikTok fun is seeing actual people do crazy stuff. Sora 2 could synthesize someone hitting five full-court shots in a row, but it wouldn’t be inspiring or engaging. How will this be different than music-generating AI like Suno, which doesn't have widespread adoption despite incredible capabilities?
It's hard to believe, but some people enjoy. On the other hand, some popular content on TikTok is probably worse than AI generated content and that's another problem...
Just seeing the examples that I assumed are cherry picked, it seems like they're still behind on Google when it comes to video generation, the physics and stylized versions of these shots seem not great. Veo3 was such a huge leap and is still ahead of many of the other large AI labs.
I can see it being interesting to create wacky fake videos of your friends for a week or two, but why would people still be using this next year?
I watch videos for two reasons. To see real things, or to consume interesting stories. These videos are not real, and the storytelling is still very limited.
A lot of realslop is fake too. As in staged but pretended as real for rage bait or annoyance bait. Or stupid shaggy dog story videos, where it seems like the thing will happen any moment now and then nothing happens.
One recent disillusionment for me was that lots of police body cam content is fake, as in basically amateur actors trying to enact a realistic police stop, they even put the usual bodycam numbers and letters and axos logo in the corner etc.
And so many other videos of things happening in the street are more or less obviously fake and staged. Still 90% probably don't notice.
> Concerns about doomscrolling, addiction, isolation, and RL-sloptimized feeds are top of mind—here is what we are doing about it.
> We are giving users the tools and optionality to be in control of what they see on the feed. Using OpenAI's existing large language models, we have developed a new class of recommender algorithms that can be instructed through natural language. We also have built-in mechanisms to periodically poll users on their wellbeing and proactively give them the option to adjust their feed.
So, nothing?
I can see this being generated and then reposted to TikTok, Meta, etc for likes and engagement.
It's insanely impressive. At the same time, all these videos all look terrible to me. Still get extreme uncanny valley and literally makes me sick to my stomach.
This stuff works really well when you make something that's exaggerated reality, as in either an animation or a MTV-style music video
I can't find the link now, but I saw a continuous shot video of a grocery store from the perspective of a fly. It was shot in the 90s music video style and looked so damn good.
Some of the stuff being done by these guys is also a whole lot of fun (slightly NSFW and political content), and it fits the music video theme:
Agree - leaps and bounds beyond anything I would have dreamed possible a few years ago...but... IDK, if I'm honest, the sound was way off too, not just the visuals. The music sounded detuned slightly, and the crowd noise was "crackly" etc. etc. It had a low-fidelity "quality" to it.
Personally, I feel mixed feelings. I'm impressed, but I'm not looking forward to the new "movies" that are going to litter YouTube et al generated from this.
Maybe by Sora 3 they will have scene consistency. Gah it's so jarring to me that the poll the racing ducks are in just randomly changes. My brain can tell it's not consistent scene to scene and feels so jank.
I saw a famous actor-director (can't remember who, but an A-list guy) said it would be super valuable even if you only use it for establishing shots.
Like you have an exterior shot of a cabin, the surrounding environment, etc — all generated. Then you jump inside which can be shot on a traditional set in a studio.
Getting that establishing shot in real life might cost $30K to find a location, get the crew there, etc. Huge boon to indie films on a budget, but being able to endlessly tweak the shot is valuable even for productions that could afford to do it IRL.
Searched around and found it. It was actually Ashton Kutcher's interview with Eric Schmidt.
Kutcher mentions the establishing shots, and I'd forgotten also points out the utility for relatively short stunt sequences.
> Why would you go out and shoot an establishing shot of a house in a television show when you could just create the establishing shot for $100? To go out and shoot it would cost you thousands of dollars.
> Action scenes of me jumping off of this building, you don’t have to have a stunt person go do it, you could just go do it [with AI].
Casey Affleck is currently shooting a horror vampire period piece using Comfy UI and an Unreal Engine Volume. The AI is used for the background plates. It's just a test, but it's happening right now.
Wow. What an intelligent take. I would have never expected this from Ben Affleck. He seems extremely familiar with the technology and it's capabilities and limits.
People use these for sure but the biggest problem with these I feel is that they produce ”finished shots” with 8 bit colors and heavy grading. It’s hard to mix it with the other material which actually looks quite bland while it is being worked on. Would be great if somebody would train a model on raw footage.
I think it will mainly be used for inpainting and outpainting, i.e. for adding and removing stuff in a scene. Things that currently have to be done with relatively expensive CGI. More complex things, especially things that have to look identical between scenes, or have to look a very specific way, will still require filming or classical CGI, or both. (For now.)
Ugh. While technically extremely impressive, I'm so tired of the slop. Every AI content generation tool should have a watermarking system in place, and sites like YouTube should have a way to filter out AI generated content from search results with the press of a button.
Ever since the launch of Veo, there's already so much AI slop videos on YouTube that it becomes hard to find real videos sometimes.
I feel that this is a data collection activity (and thus, more advanced future models and usecases) disguised as a social media. People will provide feedback in the form of clicks/views on AI generated content (better version of RLHF) on unverified/subjective domains.
Biggest problem OpenAI has is not having an immense data backbone like Meta/Google/MSFT has. I think this is step in that direction -- create a data moat which in turn will help them make better models.
Wish I was cool enough to have an invite code. Oh well, as an iOS build nerd next best thing I can do is inspect their ipa I guess. Interesting that they have some pretty big duplicate mp4s nobody caught in NoFaceDesignSystemBundle: cameo_onboarding_0.mp4 & create_ifu_1.mp4 | 7.3MB and cameo_onboarding_2.mp4 & create_ifu_0.mp4 | 5.2MB.
Also I find it neat that they still include an iOSMath bundle (in chatGPT too), makes me wonder how good their models really are at math.
in a computer graphics course i took, we looked through how popular film stories were tied to the technical achievements of that era. for example, toy story was an story born from the new found ability to render plastics effectively. similarly, the sora video seems to showcase a particular set of slow moving scenes (or when fast, disappearing into fluid water and clouds) which seem characteristic of this technology at the current moment in time
What is the target market for this? The videos are not good enough for YouTube. They are unrealistic, nauseating and dorky. Already now any YouTube video that contains a hint of "AI" attracts hundreds of scathing comments. People do not want this.
Let me guess, the ultimate market will be teenagers "creating" a Skibidi Toilet and cheap TikTok propaganda videos which promote Gazan ocean front properties.
Soon, you won't even have to do anything to post a video of yourself doing something "interesting" on social media, what at time to be alive.
There would for sure be large swathes of people who would just lie about what they're doing and use AI to make it seem like they're skateboarding, or skiing or whatever at a pro or semi-pro level and have a lot of people watch it.
From watching the video I have an impression that these guys just want to appear cool, and the product looks like that too. To appear to be very cool, for people who won’t ever use it, apparently. Same impression I’ve got from watching that promo with Jony Ive. Beautiful, and don’t you dare to think it through.
It's very hard for simultaneous good audio generation with video generation (simultaneous generation is necessary to maintain lip sync). Veo 3 et al also have flat monochannel audio, but not as bad as these Sora 2 demos.
This is super cool and fun and will almost certainly be really bad for society in loads of different ways. From the descriptions of all the guardrails they're needing to put in it seems like they know it too.
Glad to see someone is looking out for a forest, here. A diverse host of excuses have cropped up to explain away the anxiety AGI brings, and I totally understand why. Yet again, today we stare into the abyss.
Sora 2 represents significant progress towards [AGI]. In keeping with OpenAI’s mission, it is important that humanity benefits from these models as they are developed.
I wonder how exactly they reconcile the quote above with "We are concerned about late-stage AGI development becoming a competitive race without time for adequate safety precautions"...
I am not for or against AGI, but why is there anxiety around it? Do people simply hear sales rhetoric and assume that it can exist and will be used in order to dominate their lives?
I'm not referencing sales rhetoric, I'm referencing scientific consensus. AGI will have the same kind of impact on our species as fire and electricity did. We stand at a crossroads between unimaginable success and enormous catastrophe...
So being able to generate sign language videos for people who cannot hear is not a legitimate use case for AI videos? Or is your hate boner for AI just blinding you from useful applications?
Great point. Really, the main problem with subtitles is that the creator can understand them without having to know another language, and therefore can spot check them. That makes it much more difficult to insert Black Mirror-style Contextually Relevant Advertisements.
sign languages are completely different languages from spoken languages, with their own grammar etc.
subtitles can work but it's basically a second language. perhaps comparable to many countries where people speak a dialect that's very different from the "standard" written language.
this is why you sometimes have sign language interpreters at events, rather than just captions.
there's not really a widely accepted written form of sign language.
> subtitles can work but it's basically a second language
That argument applies just as equally to sign language - most countries have their own idiosyncratic sign language. (ASL, LSE, etc.). Any televised event that has interpreters will be using the national language version.
The closest thing you're thinking of is IS - International Sign but its much more limited in terms of expression and not every deaf person knows it.
> there's not really a widely accepted written form of sign language.
Because it makes no sense to have it unless there was a regional deaf community that was fluent in sign language and also simultaneously illiterate.
>this is why you sometimes have sign language interpreters at events, rather than just captions.
No, the reason is because a) it's in real time, and b) there's no screen to put the subtitles on. If it was possible to simply display subtitles on people's vision, that would be much more preferable, because writing is a form of communication more people are familiar with than sign language. For example, someone might not be deaf, but might still not be able to hear the audio, so a sign language interpreter would not help them at all, while closed captions would.
if you're maximizing accessibility you'd have both. often in broadcasts with closed captioning, there will also be a video of the sign language interpreter.
The real question is... what is the advantage of written sign language versus... normal writing? I think a lot of people are confused and think that there is only one universal form of sign language used worldwide [1].
Second problem is that sign language is heavily influenced with corresponding facial expressions, body language, the motion of the hands, even how emphatic the motions are. Trying to approximate what is effectively a SPATIAL language into written glyphs feels like a complete waste of time.
I feel like we might be talking past each other but it is funny that you chose French and Swahili. [1]
The point is that "SIGN LANGUAGE" is idiomatic to the native speaker's tongue. So if you're going to take the time to create a specialized written form of it, you can just write using the native language which can be read by BOTH the Deaf and non-Deaf community.
Deaf people are not magically illiterate.
Creating a written sign language serves no value since it is just a crappier version of the normal written equivalent.
So there's not a lot of value in creating a written form of say the French Sign Language because you can just use French.
Swahili regions have multiple types of sign language including Kenyan Sign Language.
> The point is that "SIGN LANGUAGE" is idiomatic to the native speaker's tongue.
No, this is not true. French and French Sign Language are totally unrelated languages. Sign languages generally have little to do with the spoken language of the country they’re used in, that’s why for example American Sign Language and British sign language are completely different and not mutually intelligible despite the UK and the US speaking the same language (with only slight differences in accent and vocabulary).
AI is irrelevant to the reason why there isn't a written version of every single national dialect of sign language. The reason it doesn't exist is because it would serve no purpose (source: many deaf friends). Deaf communities learn the country's writing system just like everyone else.
The closest thing out there is SignWriting [1] which has about as much traction in the real world as esperanto.
I think someone had called it many months back (and in fact I felt it too) that the feed for Sora seemed very much like a social media app. Then the only thing left was to make it into vertical scrolling with videos and voila you have your tiktok clone.
Shameless plug but I am creating a startup in this space called cleanvideo.cc to tackle some of the issues that will come with fake news videos. https://cleanvideo.cc
While the quality of what I'm seeing is very nice for AI generated content (I still can't believe it) but the fact thay they are mostly showing short clips and not a long connected consistent video makes it less impressive.
Does this survive panning the camera away for 5 to 10 seconds and then back? Or basic conversation scene with the camera cutting between being located behind either speaker once every few seconds?
Basically proper working persistence of the scene.
Dude, this generation of AI video models are just starting to have basic camera production terms understood, and then it is exactly like LLM generation: it's a pull of a slot machine arm; you might get what you want, but that's "winning" and the slot machine only gives out winners one in every 100 pulls. Every possible thing that could not be right happens.
For example, I'm working with a walking and talking character at this time using multiple AI video models and systems. Generated clips any length longer than 8 seconds risk rapid quality loss, but sometimes you can get up to 12-19 seconds without the generation breaking down. That means one needs to simulate a multiple camera shoot on a stage, so you can cut around the character(s) and create a longer sequence. But now you need to have multiple views of the same location to place your character(s) into - and current AI models can't reliably give you a "different angled views" of an environment. We just got consistent different views of characters, and it'll be another period until environments can be generally examined from any view. BUT, that's if people realize this is not in the models yet, and so far people are so fascinated by the fantasy violence and sexual content they can make nobody realizes you cannot simply "look left and right" in any of these models and that even works with consistency or reliability. There are workarounds, like creating one's entire set and environments in 3D models, for use as the backgrounds and starting frames, but that's now 3D media production + AI, and none of the AI tools generate media that even has alpha channels, and a lot of similar incompatibilities like that.
Find this sort of innovation far less interesting or exciting than the text & speech work, but it seems to be a primary driver of adoption for the median person in a way that text capability simply is not.
However, personalization (teleporting yourself into a video scene) is boring to me. At its core, it doesn't generate new experience to me. My experience is not defined by photos / videos I took on a trip.
I also can't think of a reason why I would ever want to look at an AI generated video.
however as they hint at a little in the announcement, if video generation becomes good enough at simulating physics and environments realistically, that's very interesting for robotics.
A little tangential to this announcement, but is anyone aware of any clean/ethical models for AI video or image generation (i.e. not trained on copyright work?) that are available publicly?
If I was on the OpenAI marketing team I maybe wouldn't have included the phrase "and letting your friends cast you in their [videos]". It's a little chilling.
You can use the "cameo" feature only with users who have gone through the cameo creation flow. Sama has an account and created a cameo likeness of himself.
When you create your cameo you can choose who is allowed to make videos using it: "only me", "people I approve", "mutuals", or "everyone".
So far the true progress it has made is getting textures right close up. It still fudges how skin looks like the more it pans away from the characters.
I really hope they have more granular APIs around this.
One use case I'm really excited about is simply making animated sprites and rotational transformations of artwork using these videogen models, but unlike with local open models, they never seem to expose things like depth estimation output heads, aspect ratio alteration, or other things that would actually make these useful tools beyond shortform content generation.
I just had a thought: (spoilers Expanse and Hyperion and Fire Upon the Deep)
Multiple sci-fi-fantasy tales have been written about technology getting so out of control, either through its own doing or by abuse by a malevolent controller, that society must sever itself from that technology very intentionally and permanently.
I think the idea of AGI and transhumanism is that moment for society. I think it's hard to put the genie back in the bottle because multiple adversarial powers are racing to be more powerful than the rest, but maybe the best thing for society would be if every tensor chip disintegrated the moment they came into existence.
I don't see how society is better when everyone can run their own gooner simulation and share it with videos made of their high school classmates. Or how we'll benefit from being unable to trust any photo or video we see without trusting who sends it to you, and even then doubting its veracity. Not being able to hear your spouse's voice on the phone without checking the post-quantum digital signature of their transmission for authenticity.
Society is heading to a less stable, less certain moment than any point in its history, and it is happening within our lifetime.
VFX artist and developer here, who's deep into this stuff, and it is really not there. It's an island of itself, barely controllable and barely usable with other media. They are just now getting around to generating alpha channels, with virtual none of the existing pipelines for any AI video or image generation tools to even incorporate and work with alpha channels. This is just one of several hundred aspects of incompatibility. It really seriously appears as of no one at any of the AI video generation research teams has any professional media production experience, or even bothered too look at existing media production data standards, and what they are making tool-wise is incompatible in every possible respect.
"It really seriously appears as of no one at any of the AI video generation research teams has any professional media production experience, or even bothered too look at existing media production data standards,"
I had to chuckle at this. Because the arrogance of OAI et al will finally get them in the end when these projects continue to be negative NPV.
The cost is and will be more than that, the time will be more, and I really think people are underestimating the time it takes to create good stories. Sure, there will be online locations to make short form video of all kinds. People have had video cameras in their pockets for a very long time and being hobby film makers are not really popular. The AI video sites now are 95% people fascinated with the ability to make video at all, and after a bit their interest dies because to actually make anything that requires real work even with AI helping left and right. Consistency is a harsh mistress; and AI video is only good with it for a short duration. So any narrative that makes a story worth watching, it's not AI slop, will continue to require humans and human creativity - for the consistency that gives a story the integrity that makes it worth watching. At least for audiences that care. No doubt, there are commercial forces working to develop audiences that like and prefer AI slop.
Do you even see a path from the current AI systems to something that has that near-total control over every detail that is required for high quality VFX work?
Yes. Adding alpha channels would be step one. Then perhaps incorporate the "element" concept that is basically any identifiable visual anything; which is what VFX uses as a composite-capable element. Then build a whole visual scene description prose that is what we give to a video AI, and that prose is high level language where necessary and element-wise specific where necessary. Base that scene description prose on the language used by film makers directly, just adopt their terminology, and then track the industry's jargon within the models. That way anyone working in media will auto-magically know how to control them.
We are at a point now where it is now how to write software that is the problem but how to describe to the software that is the problem. Video and film making is so generalized, AI needs more information. Typically that information comes from a director's and their team's consistency during production. AI has neither the information for consistency of imagery nor the narrative and the perspective of the narrative a human director and team bring. In time, AI will develop large enough contexts, but will the hardware to run that be affordable? There is a huge amount of context in both an entire script and the world view perspective a film crew brings to any script, and for that reason I think many of the traditional (VFX included) film roles are not going to suddenly disappear. AI video does not replace their consistency at their budget, hands down.
When AI video is able to be just a part of the skill set, for example when it is compatible with compositing, editing, and knows that terminology, AI video will be adopted more. Right now, it is designed as an all or nothing offering.
New grads are already having a tough time. My own expectation is that every recession or downturn from here on out, there will be the typical rounds of layoffs but without the typical increase in hiring afterwards. Maybe no “we replaced you with AI” moment, more of a “no new hiring” tendency.
Art should require effort. And by that I mean effort on the part of the artist. Not environmental damage. I am SO tired of non tech friends SWOONING me with some song they made in 0.3 seconds. I tell them, sarcastically, that I am indeed very impressed with their endeavors.
I know many people will disagree with me here, but I would be heart broken if it turned out someone like Nick Cave was AI generated.
And of course this goes into a philosophical debate. What does it matter if it was generated by AI?
And that's where we are heading. But for me I feel effort is required, where we are going means close to 0 effort required. Someone here said that just raises the bar for good movies. I say that mostly means we will get 1 billion movies. Most are "free" to produce and displaces the 0.0001% human made/good stuff. I dunno. Whoever had the PR machine on point got the blockbuster. Not weird, since the studio tried 300 000 000 of them at the same time.
Who the fuck wants that?
I feel like that ship in Wall-E. Let's invest in slurpies.
Anyway; AI is here and all of that, we are all embracing it. Will be interesting to see how all this ends once the fallout lands.
Sorry for a comment that feels all over the place; on the tram :)
Cool - now let's see how much it costs in compute to generate a single clip. (Also, notice how no individual scene is longer than a handful of seconds?)
This is going to be a disaster. We are never going to be able to trust a video again and in short order propagandists are going to be using this to generate god knows what.
This is the kind of thing people get excited about for the first couple of months and then barely use it going forward. It's amazing how quickly the novelty of this amazing technology wears off. You realize how necessary meaning/identity/narrative is to media and how empty it gets (regardless of the output) when those elements are missing.
So, this is the AI Slop generator for the AI SlipSlop that Altman has announced lately.
Brave new internet, where humans are not needed for any "social" media anymore, AI will generate slop for bots without any human interaction in an endless cycle.
Now videos will be generated on the fly based on your preference. You will never put your phone down, it will detect when your sad or happy and generate videos accordingly
I wonder how this will affect the large cinema production companies (Disney, WB, Universal, Sony, Paramount, 20th century...). The global film market share was estimated to be 100B in 2023. If the production cost of high FX movies like Avengers Infinity War goes down from 300M$ to just 10K$ in a couple of years, will companies like Disney restrain themselves to just release a few epic movies per year? Or will we be flooded with tons of slop? If this kind of AI content keeps getting better, how will movies sustain our attention and feel 'special'? Will people not care if an actor is AI or real?
I guess copyright is pretty much dead now that the economy relies on violating it. Too bad those of us not invested into AI still won't be able to freely trade data as we please....
The ease of creating visually titillating media, coupled with the difficultly of consistency works against the creation of narrative media. I sure hope we don't get a generation of non-narrative beautiful slop.
I think HN is too political like this tech is clearly amazing and it’s great they shipped it there should be more props even if it’s a billion dollar company.
Yeah I suppose so, most comments on this kinda thing are not really discussing the technology in a vacuum. I imagine it's due to the quite cynical nature of HN at this particular time period where society is fundamentally shifting, in arguably a negative direction, with this kind of technology as one of the main reasons.
I haven't been on HN for that long, what was it like 5-10 years ago? I'm curious how it will be in 5-10 years.
Yes the tech is amazing. But tech is not everything, after 20 years of social media, its pretty clear to everyone that those things can have large long term impact both positive and negative for society, discussing the potential impacts of the tech is not being "political", its just being interested in the future.
Well certainly if you don’t want us to discuss the possible implications ahead, then yes we can only close our eyes and learn from hardship once it’s there, but then what do we learn ? To not close our eyes next time ? We could just do that like now.
I think it depends on the time and place. So in this context it doesnt matter too much, its a billion dollar company, but if it was a guy with a project he just spent all year on, publishing it on HN for the first time, I would expect people to focus less on teh politics and more on the achievement which is something I dont see too often, unless the work is stellar. Perhaps there is just a high bar on HN
“With Sora 2, we are jumping straight to what we think may be the GPT‑3.5 moment for video.”
I think feeling like you need to use that in marketing copy is a pretty good clue in itself both that its not, and that you don’t believe it is so much as desperately wish it would be.
I am looking at the videos and really had a feeling that it looks right (minus a lot of obvious fuck ups still) where previously something felt fundamentally wrong with ai videos. It feels somewhat important, in so far you consider ai generated videos important.
It's the skin textures. It's the slightly better lipsyncing. Maybe it will be different when us normal users get it but so far the demos with Sam don't make him look waxy.
The Sora app squaring off against Meta's social video app is the real story here.
Sora 2 itself looks and sounds a little poorer than Google Veo 3. (Which is itself not currently ranked as the top video model. The Chinese models are dominating.)
I think Google, with their massive YouTube data set, is ultimately going to win this game. They have all the data and infrastructure in the world to build best-in-class video models, and they're just getting started.
The social battle will be something completely different, though. And that's something that I think OpenAI stands a good chance at winning.
Edit: Most companies that are confident of their image or video models stealthily launch it on the Model Arena a week ahead of the public model release. OpenAI did not arrange to do that for Sora 2.
Nano Banana, Seedream/Seedance, Kling, and several other models have followed this pattern of "stealth ELO ranking, then reveal pole position".
The fact that this model is about "friends" and "social" implies that this is an underpowered model. You probably saw a cherry picked highlight reel with a large VRAM context, but the actual consumer product will be engineered for efficiency. Built to sustain a high volume of cheap generations, not expensive high quality ones. A product built to face off against Meta. That model compete on the basis of putting you into videos with Pikachu, Mario, and Goku.
> I think Google, with their massive YouTube data set, is ultimately going to win this game.
I don't know, applying the same thinking to LLMs, Google should have been first and best with just text based LLMs too, considering the datasets they sit on (and researchers, among others the people who came up with attention). But OpenAI somehow beat them on that regardless.
The problem for Google existed with the infobox at the top of search results. If users get the answer to their query without having to visit the web page where the answer came from, and where Google shows the ads, means that users don't see ads, and that website operators don't get ad revenue. ChatGPT was Google's Kodak digital camera moment. They had internal transformers-based chatbots (that really wanted to send you pizza,
for some reason), but deploying that would have cannibalized their existing business model, so in the meanwhile, their lunch got eaten by an outside competitor.
As usual: impressive until you look close. Just freeze the frame and you see all the typical slop errors: pretty much any kind of writing is a garbled mess (look at the camera in the beginning). The horn of the unicorn sits on the bridle. The buttons on Sam's circus uniform hover in the air. There are candleholders with somehow candles inside as well as on top. The miniature instruments often make no sense. The conductor has 4 fingers on one hand and 5 on the other. The cheers of the audience is basically brown noise. Nedless to say, if you freeze the audience, hands are literally all over the place. Of course, everything conveniently has a ton of motion blur so you cannot see any detail.
I know, I know. Most people don't care. How exciting.
Is your complaint that it has errors? I mean look at what it can do. This is a freaking computer generating things from scratch based on a prompt. Two years ago, technology like this was so much worse and could only generate basic images and videos. Now it can generate visuals all from the text someone puts in.
Anyone, literally anyone, can use it (eventually) to generate incredible scenes. Imagine the person who comes up with a short film about an epic battle between griffins and aliens...Or a simple story of a boy walking in the woods with their dog...Or a story of a first kiss. Previously people were limited to what they had at hand. They couldn't produce a video because it was too costly. Now they can craft a video to meet their vision.
Well, yes? There's a reason why everything that was produced with these tools so far is garbage: because no one actually caring about their art would accept these things. Art is a deliberate thing, it takes effort. These tools are fine for company training videos and TikToks. Of course a few years ago this was science fiction. They are immensely impressive from a technical perspective. Two things can be true.
I don't know if it's just me or other people are feeling it as well. I don't enjoy videos anymore (unless live sports). I don't enjoy reading on my monitor anymore, I have been going back to physical books more often. I am in my early thirties.
The point is that sora2 demo videos seemed impressive but I just didn't feel any real excitement. I am not sure who this is really helping.
Personally I can't wait for super creative and novel indie film productions as film production will be more liberated from the grip of Hollywood and the influence of the upper classes in general. Especially once the Chinese make less-censored-to-Western-users models more available and even more so once people can run these things at home in some years.
So a social network that's 100% your friends doing silly AI things?
I feel like this is the ultimate extension of "it feels like my feed is just the artificial version of what's happening my friends and doesn't really tell me anything about how they're actually faring."
Social media also tends to highlight the best parts of people’s lives, creating unrealistic expectations and views for those consuming it and looking at their real life. Now social media won’t even be a highlight reel, but completely fabricated.
I have to imagine there will be a rebellion against all of this at some point, when people simply can’t take the false realities anymore. What is the alternative? Ready Player One? The Matrix? Wall-E?
Maybe inside of a social network specially for AI, but a concerning number of people don't realize images and videos are AI, even when it's bad AI. As it gets better, and starts integrating the poster's image (like Sora 2), that's going to get even worse.
Some people use filters/photoshop to artificially juice their images; now they can use AI to artificially juice every aspect of their on-line presence.
I just watched the announcement video and something about it just gives me the ick. The whole time I just had the uncanny valley feeling.
The technology itself is super impressive, but a social media app of AI slop doesn't feel like the best use of it. I'm old enough to not really be interested in social media in general anymore, so maybe I'm just out of touch, but I just can't see this catching on. It feels like the type of thing that people will download, use a few times until the novelty wears off and then never open again.
I built an MVP of this [1] with images (not video) and in more of an Instagram style (not tiktok) back in ‘22, with the tagline ‘What if social media were literally fake?’
I am bullish on this, albeit with major concerns in many domains. It was fun and addictive as hell with images. With video it will be wild.
I find it comical that OpenAI with all the power of CharGPT even them are unable to release an app for both iOS and Android at the same time. Wow, good marketing for Codex.
Or Sama's documented reverence for Apple products. We are talking about the guy who sold Tim Cook his AI for $0.00, he's not exactly got the horse drawing the cart here.
Overall, appears rather underwhelming. Long way to go still for video generation.
Also, launching this as a social app seems like yet another desperate try to productize and monetize their tech, but this is the position big VC money forces you into.
I've perhaps been away from the scene for a bit, but I'm very impressed. To me this is absolutely "video generation", and I don't get your disdain for productization and monetization; last I checked this wasn't "Basic Research News".
I don't think it's disdainful to point out the lack of PMF for a dedicated app for Sora, nor how its behind competitors who don't require a dedicated social app. No need to strawman the guy, I think it's okay to be reasonably critical of ideas still on this website.
Inb4 make your own video model and see how easy it is
Every AI video demonstration is always about funny stuff and fancy situations. We never see videos on art, history, literature, poetry, religion (imagine building a video about the moment Jesus was born) ... ducks in a race !? Come on ...
So much visual power, yet so little soul power. We are dying.
What do you imagine a generated video about poetry would be?
>Every AI video demonstration is always about funny stuff and fancy situations.
The thing about AI slop is that by its very nature, unless it's heavily reined in by a human, it's invariably lowest common denominator garbage. It very likely will generate something you yourself could think of within the first five seconds of hearing the prompt, not some very clever take on it, so it can only work as a placeholder (AI as a replacement of stock images is great, for example) or to add background detail where it won't call attention to itself and its genericity.
>imagine building a video about the moment Jesus was born
Given there are multiple paintings on the subject, I very much doubt no one has generated something like that already.
After the disaster that was chatGPT4.001, study mode and now this: an impossibly expensive to maintain AI video slop copyright violater, their releases are uninspired and bland, and smelling of desperation.
Before AI, each video on this channel would have taken a large team with a Hollywood budget to create. In a few more years, one person may be able to turn their creative vision into a full-length movie.
No, but there's some stuff that are really creative. Ironically I think the reason I'm more positive about it is because I only encounter AI generated (non-text media) ~ once a week / 2 weeks.
But modern AI could create images which are basically indistinguishable from a real Monet if you are not an expert. So the fact that you like Monet's pictures, but not Monet-like AI pictures, shows that part of what you like is the fact that an image is made by a specific human instead of being generated by a diffusion model.
I dunno, look at these [0] I think they're quite nice! But I can imagine getting bombarded with them all day can eventually turn someone off. (I assume I'd feel the same if I saw 100 Monet's every day for 3 years)
The value will shift to search or curation - if the cost to produce drops to nil, then the value will be in finding good content amongst a flood of sameness.
Those videos look like some teenager thoughtlessly applying an aftereffects filter(whatever) to 1000 short selfie videos. On What planet would this require a Hollywood budget and years? Who are you shilling for exactly? Do you really believe what you write.
Things are cool because they are unique, very hard to create, and require creativity. When those things become cheap commodities, they are no longer cool.
Exactly. Pushing a photo through a Van Gogh filter doesn't get near what a real Van Gogh expresses. It's in a temporal context, communicates something about the person and their thoughts about reality. Their artistic choices matter, because they can't just put out 10 different variations, instead they have to pick one. And then we can think about why that one was chosen.
The same could be said about software, and it's safe to say that open-source software making complex workflows easier and more efficient is a net good.
Making better tools is better for everyone: the median usage of those tools downstream is a separate issue.
Indeed. Art is partially evaluated by how impressive it is. That's why posting AI images on social media won't yield a lot of likes anymore. People have gotten used to images being easy to create, so they aren't seen as valuable anymore. The same will be true for videos.
AI pictures today are much less impressive than Dall-E 2 pictures were a few years ago, despite the fact that the models are much better nowadays. Currently AI videos can still be impressive, but this will quickly become a thing of the past.
Then people will move from trying to create art to creating "content". That is, non-artistic slop. Advertisements. Porn. Meme jokes. Click bait. Rage bait. Propaganda. Etc.
I would argue that we just get pickier and more sensitive to slop. When everyone can make a movie, the standard for a good movie will be higher. Many current Hollywood films wouldn’t make the cut. But maybe some kid in Nigeria makes the greatest film of all time.
Hard to interpret that comment as anything but racist. Chinua Achebe is widely considered one of the greatest modern novelists. He was 28 when he wrote Things Fall Apart.
Perhaps learn the meaning of the phrase "by commonly accepted measures" before you accuse someone of racism. I'm pretty sure hardly anyone knows about Chinua Achebe, so your definition of "widely" must be quite wide.
Things Fall Apart has sold over 20 million copies and has been translated into more than 50 languages. It is a staple of literature curriculums in schools and universities across the globe. That isn't a "wide" definition of widely known; it's the standard one.
Then you have Chimamanda Adichie, who has sold millions of copies and won several awards, including the BBC National Short Story Award, widely described as "one of the most prestigious awards for a single short story"
Then another Nigerian writer, Wole Soyinka, won the Nobel fucking Prize in Literature in 1986. Or is that measure not good enough for you, your highness ?
Not only do you come across as racist, you clearly have no idea what you're talking about. Congratulations.
These examples seem highly cherry-picked. If you look at bestseller lists, or writers who average people actually know, the results are in fact very different. Your accusation ("racist") is defamatory.
Calling a Nobel Prize winner, among others 'cherry-picked' in an argument about literary greats where you asked for 'commonly accepted measures' is one of the most intellectually dishonest things I've ever read, so congratulations again.
You were thoroughly proven wrong so now your new standard for literary greatness is "writers that average people know" ? (which is really just code for 'writers I know', because millions do know those writers, I wasn't sharing some secret). I guess that means we can throw out Faulkner, Joyce, and Woolf in favor of whoever's currently at the top of the airport bookstore list.
It's not "defamatory" to point out that your argument, which began with a dismissive generalization about an entire country, was based on profound ignorance (the kind that wouldn't have taken anything more than a basic google search to remedy). You were corrected with facts. Instead of going, 'I stand corrected, sorry', you're doubling down. It just makes you look worse, and stupid.
This is the most basic racist playbook happening in real time, and you're the star. If you genuinely think you aren't then you need to take a long, good look at yourself.
I wasn't claiming they had ownership of that aesthetic, and that sort of gets into philosophical questions about whether one can own such a thing anyways. I like the style and I'm glad they have the tools to bring it into clarity from the abstract.
Democracy? Strengthened! Nothing says “informed electorate” like not knowing if a politician actually said they support nazism or if it was just a hyper-realistic AI puppet.
Trust in media? Soaring! Why believe your eyes or ears when you can doubt everything equally?
Journalism? Thriving! Reporters now get to spend their days playing forensic video detective instead of, you know, reporting news.
Social harmony? Better than ever! Nothing brings people together like shared paranoia and the collective shrug of “I guess truth is dead now.”
lol i wonder if this will create a market for PKI at the image sensor level so that videos will be cryptographically signed and baked into the actual video stream with steganography.
Advertising: you (her) wearing new clothing before purchase, hair/glasses/makeup, make overs; guys after 3 months of gym membership, you driving the new car, you in this specific new home... etc, etc... I'm surprised this is not already everywhere, but people are too occupied making nsfw and fantasy violence clips.
Targeted advertising has become just manipulation. I don't know if personalized advertisement videos for everyone promoting a fake world that doesn't exist is really a benefit for the world...
If course it's not a benefit, but it's an advertising angle that will work very well with a class of gullible consumers, and that is enough to justify it being plastered everywhere. I don't write these rules, I just notice them.
I imagine it's incredibly useful for prototyping movies, tv, commercials before going to the final version. CGI will probably get way cheaper too with some hybrid approach.
Obviously this will get used for a lot of evil or bad as well
I feel like that's missing the point of pre-vis anyway, its purpose is to lay down key details with precision but without regard for fidelity (e.g. https://youtu.be/KMMeHPGV5VE), a system with high fidelity but very loose control is the exact opposite of what they want.
... how dare you, sir. That is entirely unacceptable and you will be reported to the ministry of proper living!
Regardless of the slop, some people will learn to use it well. You have stuff like NeuralViz - quite the sight! - and other creators will follow suit, and figure out how to use the new tools to produce content that's worth engaging with. Bigfoot vlogs and dinosaur chase scenes, all that stuff is mostly just fun.
People like to play. Let them play. This stuff looks fun, and beats Sora 1 by a long shot.
Could you please stop posting unsubstantive comments and flamebait? You've unfortunately been doing it repeatedly. It's not what this site is for, and destroys what it is for.
If Sora 2 is aiming for AI‑Tok, ScaryStories Live is the jump-scare cousin: real‑time POV horror from a photo + a sentence. No film school, no GPU farm—just “upload face, pick fear level, go.” It’s less cinema, more haunted mirror, and it ships in seconds. scarystories.live
I haven't seen comments regarding a big factor here:
It seems like OpenAI is trying to turn Sora into a social network - TikTok but AI.
The webapp is heavily geared towards consumption, with a feed as the entry point, liking and commenting for posts, and user profiles having a prominent role.
The creation aspect seems about as important as on Instagram, TikTok etc - easily available, but not the primary focus.
Generated videos are very short, with minimal controls. The only selectable option is picking between landscape and portrait mode.
There is no mention or attempt to move towards long form videos, storylines, advanced editing/controls/etc, like others in this space (eg Google Flow).
Seems like they want to turn this into AITok.
Edit: regarding accurate physics ... check out these two videos below...
To be fair, Veo fails miserably with those prompts also.
https://sora.chatgpt.com/p/s_68dc32c7ddb081919e0f38d8e006163...
https://sora.chatgpt.com/p/s_68dc3339c26881918e45f61d9312e95...
Veo:
https://veo-balldrop.wasmer.app/ballroll.mp4
https://veo-balldrop.wasmer.app/balldrop.mp4
Couldn't help but mock them a little, here is a bit of fun... the prompt adherence is pretty good, at least.
NOTE: there are plenty of quite impressive videos being posted, and a lot of horrible ones also.
Not to be a downer, but even as someone very optimistic about technology and AI generally, "TikTok but AI" sounds like a societally terrible thing to try and create.
What's the benefit of this? Curious if anyone has a solid viewpoint steelmanning any positives they can think of.
As a social experiment to reveal how senseless and pointless pop entertainment could be.
(personal rant) I've been in a mild existential crisis since I read Amusing Ourselves to Death. Can one form of entertainment really be more well-regarded than another? Is fine art fundamentally different from pop art? Are there 'finer' pop cultures amongst all pop cultures? I do still think reading The Song of Ice and Fire is more meaningful than scrolling TikTok. The crisis part is that I can't justify this belief with words.
There are two completely distinct differences that jump out to me initially that I think may help justify your feelings:
1: Reading a long book demands focus on a longer timespan than scrolling TikTok, and with focusing on a single thing for a long time, we get a sense of accomplishment. I don’t know how to justify this as valuable, but for some reason I feel that it is.
2: The Song of Ice and Fire (and GoT) were consumed by a huge proportion of people, and you now have this in common with them. This act of consuming entertainment also grants you a way to connect with other humans - you have so much to talk about. Contrast that with an algorithmic feed, which is unique just for you - no one else sees your exact feed. Of course, there are tons of people that see some of the same snippets of content, if their interests overlap with yours, but it’s not nearly as universal as having read the same series of books (and there’s much less to talk about when you’ve seen the same 17-second short form video than when you’ve both invested dozens of hours in reading the same series of books).
I don’t think these thoughts fully justify your belief, but hopefully they provide some support to it.
I think the point 2 will rub many people the wrong way (me included) though. That would make reading Fourth Wing or Twilight a more connecting experience than most classics. (Nothing inherently wrong with that, but... you know...)
The classics were classic because they were the most available and the most popular stories of their time, and they meant more in an era where creating and disseminating media was difficult. I love to romanticize a world where we go back to the classics to connect with our past and present better, even if just for the sake of efficiency. For better or for worse media is more ephemeral which means getting to a common vocabulary is one step removed. It's really a fun time to be alive.
The thing is that literature, and art in general, should be more than just entertainment. It should edify the reader, communicate some concept, moral lesson or keen insight about the world.
Remember when you were taught to extract the "moral of the story" in school? That was the whole point. That form of communication is what makes art valuable and it definitely is what makes some art more valuable than others.
Welcome to the future, where the notion of "classics" is just a point in the memetic information manifold:
https://x.com/theo/status/1973167911419412985 (Music video with Sam Altman as Skibidi Toilet)
This is pretty fun.
These keep getting wilder and wilder:
https://x.com/MatthewBerman/status/1973115097339011225 (Kinda gross)
https://x.com/cloud11665/status/1973115723309515092 (Japanese)
It can do cartoons:
https://x.com/venturetwins/status/1973158674899280077 (Rick and Morty)
https://x.com/TheJasonRink/status/1973163915476611314 (Family Guy)
https://x.com/cfryant/status/1973162037305024650 (Family Guy Horror)
Incredibly convincing anime:
https://x.com/fofrAI/status/1973164820863262748
Minecraft meets GTA:
https://x.com/Angaisb_/status/1973160337752121435
Super Mario in the real world:
https://x.com/skirano/status/1973184329619743217
Super solid looking movie trailer:
https://x.com/jasonjoyride/status/1973142061114335447
Damn:
https://x.com/theo/status/1973210960522559746
If you think this stuff is going to last longer than four months, dog, we're cooked.
I've been watching these videos for about an hour now.
I really want to call this the "Suno moment" for AI video.
Prior to Sora 2, you had to prompt a lot of clips which you then edited together. You had to create a starting frame, maybe do some editing. Roll the dice a lot.
Veo 3 gave us the first glimpse of a complex ensemble clip with multiple actors talking in a typically social media or standup comedy fashion. But it was still just an ingredient for some larger composition, and it was missing a lot of the soul that a story with a beginning-middle-end structure has.
Sora 2 has some internal storytelling mechanic. I'm not sure what they did, but it understands narrative structure and puts videos into an arc. You see the characters change over the course of the video. They're not just animated Harry Potter portraits. They're alive. And they do things that change the world they're in.
Furthermore, Sora 2 has really good "taste" and "aesthetic", if that makes sense. It has good understanding of shot types, good compositions, good editing, good audio. It does music. It brings together so much complexity in choice and arranges them into a very good final output.
I'm actually quite blown away by this.
Just like Suno made AI music simple and easy - it handled lyrics, chorus, beat, medley, etc. - this model handles all of the ingredients of a 10 second video. It's stunning.
Sora 2 isn't the highest quality video model. It doesn't have the best animation. But it's the best content machine I've ever seen.
I can see this, it's extremely impressive from a technological standpoint, and I've already been caught by the first convincing fakes on Reddit (an army person giving an anti-Trump speech). But I'm also worried, as it's a super easy channel to create convincing fakes, mass produced 'content' for mass consumption, etc.
Now these things aren't new, fake videos / images go back decades if not a century. But they took some effort to make, whereas this technology makes it possible for it to take less effort than it took for me to write this comment.
Of course, it's always my choice; if I stop visiting Reddit and touch grass instead it really won't affect me directly.
Maybe some MAY end up in compiliations in ten years, much like Vines do today. But there will be a million times more tiktoks and a billion times more AI generated videos than there were vines, so if 0.01% of vines became memetic, the amount of AI generated ones will be infintesimal.
Content is all ephemeral on some time scale, but you can cache the near-term content to maximize the views and cut back on compute costs. Some model or human made it (the cost), it's trending (the value), so keep it around for a bit.
Everything has a relevancy and penetration decay curve.
The funny thing is, I think this law applied in the classical era (1950's, 1990's, etc.), we just weren't creating at scale to realize it.
Maybe it's just one dominant variable: novelty. I'd be curious to see how we might model this.
the final one is not AI, it’s a glorb video from years ago: https://m.youtube.com/watch?v=NkYSK-_hVDQ
> Super solid looking movie trailer:
> https://x.com/jasonjoyride/status/1973142061114335447
This isn't AI generated. They're a production company and they made a short film: https://www.youtube.com/watch?v=JGLoTjxd-Ss
I think that short film is AI generated. I only watched like 30 seconds of an office scene in the middle but it spontaneously changed from daytime to nighttime with zero explanation.
He says it's not: https:/x.com/jasonjoyride/status/1973164183798816773
>> How do you get HD renders? im getting like super low res shit
>It's because this isn't AI
I haven't watched the film, but the premise is something about an orbiting space station. I could easily imagine scenes featuring rapid day/night cycles like astronauts experience on the ISS.
That movie trailer isn’t made with Sora (or AI at all, as far as I can tell?)
The irony is not lost I hope.
you are right, its an actual movie called Planet
If you've read the classics, then you will likely find a circle you can connect too. I've gone through "The Malazan Book of the Fallen" and it's a signal to know who are truly in epic military fantasy.
> That would make reading Fourth Wing or Twilight a more connecting experience than most classics.
I prefer classics myself, but this is exactly why booktok works (and why Fourth Wing blew up the way it did).
Reading the classics, in some sense, connects you to everyone who ever read them across all of human history. That's not nothing.
Depends if you are trying to connect to your contemporaries or to mankind in general. Aren't "classics" just timeless pop?
You're missing what I think is the major one: fulfillment.
> Can one form of entertainment really be more well-regarded than another? Is fine art fundamentally different from pop art?
It depends on what you want to get out of art.
Do you want human connection and shared cultural context so you can talk to real friends about things? Do you want virtual friends and connections? Do you want ideas to inspire you to create your own things, or change how you think?
Do you just want to distract yourself from how hungry you are, how much inequality is in the world, and how depressed you are, letting death draw closer?
All of those are valid things, and different art is more meaningful for different goals.
Scrolling tiktok fits into the last one, it's burning time to avoid thinking about things, moving you closer to death. Song of Ice and Fire builds a large coherent world, has bits of morality and human relation, and all of those can spark ideas and be related to your own human suffering, so it indeed feels more valid to me as a way to reflect and change how you think.
I think reading does force more long term focus, even if it's marginal for certain books. Certainly moreso than scrolling TikTok.
My personal process of grappling with this led to a focus on agency and intentionality when defining the difference.
Scrolling TikTok, much as scrolling Twitter or Facebook or Instagram or YouTube's recommendations would be, is an entirely passive activity. You sit back and you allow the Content to be fed to you.
Reading a book requires at least a bare minimum of selecting a book to read, choosing to finish that book, and intentionally choosing at any given time to spend your time reading that particular book. Similar things can be said for selecting movies. The important part in my mind is that you chose it, rather than letting someone or something else pick what they think you'll like.
The process of picking things yourself allows you to develop taste and understand what you like and dislike, mentally offloading that to someone or something else removes the opportunity to develop that capability.
I think there's arguments to be made against this view: how can you decide what to read or watch without getting recommendations or opinions? If you only engage with popular media isn't it just a slower process of the same issue?
But I do believe there is a fundamental difference between passivity and active evaluation of engagement as mental processes, and it's the exact reason why it is harder to do than scrolling is.
Where does HN comment lurking lie in the range between passivity and active evaluation, I wonder?
Eh, old people always complain about the media of the younger generation.
Compare https://de.wikipedia.org/wiki/Lesesucht (use Google Translate).
I think this is a pretty lazy dismissal as far as things go. Yes people "always complain" about many things, but that's the correct response to things that are always getting worse.
The gist of your linked article is that they were opposed to reading because they believed that reading distracted people from labor, which they considered undisciplined and immoral. Of course there also seems to be a healthy dose of misogyny associated with it:
> Poeckel's statement that women should acquire a certain amount of knowledge, but not too much, because then they could become a "burden on human society," is representative of many other texts in which reading regulations played a central role.
Then once you get to the progression of books > comics > movies > Youtube > TikTok (did I miss any?), you can observe a steady decrease in the amount of cognitive effort required to engage with the medium and a reduction in attention spans. Reduced attention span is a legitimate concern and it's only getting worse as time goes on (ask teachers).
I actually enjoy TikTok in moderation these days but I worry about people who struggle to engage with anything but TikTok, it's like a generational ratchet that only seems to go one way, towards shorter and shorter attention spans.
Maybe someone can make the argument that this won't actually matter, but it's incorrect to say that things haven't changed in observable and measurable ways, and that people are just complaining about nothing.
Am I that old already? I just turned 29 a few months ago.
While I believe that long form content such as YouTube essays can actually be intellectually stimulating depending on how you engage with the video itself (e.g. do you actively watch it, or do you just have it on in the background?), I truly believe that 95% of TikTok is just mindless slop.
My S.O. probably spends 3 hours a day on TikTok/Reels and I seriously doubt they could remember even 10% of what they saw in that time. It's like a part of their brain turns off while scrolling.
The TikTok feed is an amalgamation of posts from lots of people who aren't collaborating. The Song of Ice and Fire is a single work by a single author (or so I assume). So it's more like you're reading a single humongous post that has been “liked” (bought, positively reviewed, critically appraised) by a shitton of people, compared to a firehose of morsels that barely anyone cares about.
I’ve been in an existential crisis after reading Postman and I’ve since reframed the whole dilemma thusly: one of the highest aspirations for a person is the act of creation, and the result one can often call art.
What is wrong is instead the routine consumption of art created by others in a stupor to rest from the drudgery of daily work.
Create art, don’t waste your life consuming.
Could it be the incentives? With regard to books, paintings, theater, etc. you have an incentive to produce something that is meaningful or at least entertaining. Generally the artist is attempting to turn abstract thoughts or ideas into something real or quantifiable.
But TV and Social Media have their incentives twisted. It's just about ads. They don't really care what you are seeing as long as you are seeing as many ads as possible. The joke about TV was that a valid description of it was advertisements with a little bit of entertainment sprinkled in throughout.
I'm not saying that people haven't been able to use these platforms to build anything meaningful, but that the incentives and the purpose of these platforms are not to entertain, but to keep you glued to the feed for as long as possible to see as many ads as possible (which is why I think "rage bait" is so common).
I think some forms of entertainment can have also redeeming qualities. A novel can be seen (also) as a form of entertainment but it can also be a vehicle for a message. The difference with social media sized alternatives is that with the latter the "consumer" is much more passive, at most it's expected to react emotionally without thinking. On the other hand with the former there is an interaction between the work and the reader/viewer. Some books have the ability to make you re-evaluate your beliefs and your values, without being manipulative. Art is not necessarily entertaining.
One analogy is to liken tiktok (and shortform content) as exploring the shallows. Walking around, close to the shoreline, you explore pieces of flotsam that the sea washes your way. You might spend a lifetime on this shore, walking up and down, but most would argue that you've actually never gone anywhere.
On the other hand, reading a book is like getting on a boat. You've made certain preparations for acquiring the vessel and set course through unknown territory. A journey away from the shore and away from what's immediately at hand, which can also turn out to be a journey towards self-discovery.
> The crisis part is that I can't justify this belief with words.
Here's one attempt; it's art versus content. Tiktok is content; it's people recording a video, sometimes in one take and publishing, sometimes in multiple takes with some editing etc, sometimes fully professional ones. But overall, it's cheap, rapidly produced content for cheap, rapid consumption. ASoIaF was a labor of years to produce not just a series of books, but a world, a rich history, and later on a multi-media enterprise that involved and employed millions of people, then entertained and excited hundreds of millions of people over the years.
AI is lowering the barrier to entry even more, with anyone able to just punch in some words - less even than this comment - and produce something. For someone to consume. Maybe one in a billion will be remembered or still popular in a decade (like how some of these cheap videos are still popular / remembered / quoted, think vines / memes). But the ratio just keeps getting worse.
ASoIaF to a TikTok video is like... ASoIaF to a tweet.
A good fiction novel has multiple aspects: of course entertainment/escapism, but also a larger point the author wants to explore. With Asoiaf George Martin wanted to break/subvert classic fantasy tropes. For example that the good guy wins (Rob Stark is marrying for love and punished for that in the red wedding) or the romantic knightly quest, here done by Brianne of Tarth, an ugly/strong woman instead of a male fighter.
I am not sure how important fiction novels are (compared to reading non-fiction books or biographies who tell true facts about the real world), but I would say they broaden the horizon of the reader? And there is a selection effect in that “literature” was done by pretty smart people.
Scrolling TikTok is often described as mindless and with people not describing later what videos they watched. In general short form content (TikTok, Instagram, X/Tweets) seem to be much more superficial than long form content (eg this hn discussion board).
There's a fitting quote from 2017's Columbus:
> "[...] in its place, he identifies a different kind of crisis. Not the crisis of attention, but the crisis of interest."
Our attention in fact, has never been as fully absorbed as is today's. In place of books and architecture (as in the film), our attention has shifted towards more rapid forms. Yet in terms of hours spent, our 'attention' towards them has massively increased.
Is the crisis we're feeling then one of purported inattention, or a general loss of interest and satisfaction from our surrounds? What has spurred this crisis? Gabriel and Casey's conversation ends:
> "What about everyday life? Are we losing interest in everyday life?"
The film offers an hopeful answer.
I prescribe Plato - Republic, book X, specifically. How can one set of shadows on the wall be better than another, as they’re just shadows, degraded representations of the real?
Or perhaps Aristotle’s Poetics - pop culture has value because it is mimetic, and AI generated pop culture is no less a mirror, just one which produces reflections of every moment, all the time - but rather than the grand catharsis we might experience in a work of literature with well wrought characters with whom we empathise, we find the void staring into us as we do into it. Hollow art for hollow men.
Like it or not, the void is culture, and has value because it reflects us, albeit through a glass, darkly.
One path that might help you work out your own personal justifications is to find two forms of entertainment you enjoy at near equal levels, but where one you view as valuable and another as a waste. Then look at how both impact your life and see if you can identify what makes one valuable and the other a waste. This not only gives you a good inside view of what is happening with both forms of entertainment, but removes any bias to see your own version as superior because both forms of entertainment belong to you.
I did this, found two things I did for fun, both consuming significant blocks of time. The one that felt useless left not real impact. I want to do more of it, but after spending hours on it, I'm no different than I was before (other than perhaps a bit more skilled at the form of entertainment).
The other form, which was the same thing from an outside perspective (for example, my parents would see them as the same) left me different. It led to me building new goals, reevaluating things happening around me, spend more time thinking about where I'll be in 10/20 years. It led to me walking an hour a day and to start jogging some to build my endurance, despite the form of entertainment being unrelated to physical activity. I don't think this is innately a property of one entertainment form over another, but more about my personal relationship to entertainment.
Using this, how do 'poorly regarded' entertainment impact those engaging in it, compared to 'well regarded' entertainment? Are their lives better for it?
An old welder once told me: "It is not so important what you do. A bit more important is how you do it. And most important is why you do it."
Your comment makes me think that we have criminalized and squashed entertaining but obviously political writing out of existence :-) .
Say that somebody writes to make certain ideas more visible. For example, somebody wants people to buy the idea that amusing oneself to death is what we do (the book you mentioned). Somebody else perhaps has found that we are chronically depressed and cynic, when instead we should be thinking that a dead death itself is a fine trophy to hang on the wall during the march of progress[^1].
You can a) decide that you are set on your ways, thus entertainment should be pure and removed enough from reality so it doesn't mess with your deeply held beliefs and not read any of those books. or b) run the risk and read the thing with an open mind.
A lot of people are in the a) camp. Those who are in the b) camp would still like to be entertained a little.
[^1] Yours truly. I do that in fiction. https://www.ouzu.im/
There are probably ways you could explore this quantitatively by trying to measure the amount of novel latent information in the data you are ingesting, or by trying to quantify its effects on cognition.
Most short form content would probably score low. It’s short, for one, and it tends to be repetitive and lack anything like plot complexity or nuance.
Of course it’s not like trite pop is new. Way back in the dime store novel days it was called pulp. TikTok is just one of the latest iterations. People have always consumed dumb filler.
I read AOtD and I had a different takeaway: TV is great entertainment (watch all the trash you want!) but it's terrible for news and learning about the world. I found it pretty convincing--it's so clear to me that our societal discourse divebombed when TV news became dominant.
>The crisis part is that I can't justify this belief with words
Reading thousand page novels requires actively engaging with the material as you grow your vocabulary, and explore new ideas.
Scrolling TikTok on the other hand is a passive process. Could you recall even a quarter of all the videos you see on your TikTok feed in a single day? I would doubt it.
If society only consisted of the people in a given sector/industry, could it continue and flourish? If we only had engineers, how would society fare versus if we only had influencers? In this paradigm, there's no difference between fine art and pop art.
> The Song of Ice and Fire
is a trash derivative "we have Lords of the Rings at home" wannabe, completely void of joy and feels like it's written by an angsty edgy teenager who hates the world and has learned about medieval history for the first time and wanted to add zombies and dragons, the most original fantasy tropes.
I would honestly and unsarcastically take a day of scrolling through TikTok over sitting through 1 chapter of ASoIaF.
And apparently lately the author feels so too.
that's a really great question.
I think that it reduces down to "reward without effort is bad for you" - in so many different contexts in life, especially entertainment.
> Can one form of entertainment really be more well-regarded than another?
This is a no-brainer question. For an extreme example: CSAM is a form of entertainment for some people.
well, one destroys your attention span and brain
You might enjoy some of my writings on formalizing meaning (see here[1] and follow the links). In some way, although not always reliable, you can say that if you feel A is more meaningful than B, that is already some kind of evidence for this assertion even if perhaps unreliable in some ways.
So there isn't necessarily some huge crisis that you need to justify: in some ways reality just is (and this includes subjective reality;).
Say if you ask why do the laws of physics conserve energy locally, you can actually argue that if it were otherwise actually life would be extremely more unlikely, as that tends to increase instability in various systems (both energy divergence and going to 0 makes life unlikely); but still I'm almost certain you could conceive of forms of life in non-energy-conservative systems (something like Conway's Game of Life, but maybe with more advance rules if you prefer). So while it makes sense that the physics in our universe is approximately locally conservative (maybe not exactly in GR?), in totality it's just kind of a brute fact, an experimental observation. Our theories help us devise say better experiments to test e. conservation, and in a way map out the landscape of consistent physical laws. But they don't tell you which realization of consistent or admissible laws you'll find yourself in.
Other way to phrase it, what you feel is in a way real. So if you feel in some fundamental way better reading A than B, then that simply reflects a property of reality and needs no further explanation. The only problem is that in some cases our judgement can be distorted, like by substances or maybe overwhelming blinding desires (that fail to reflect fundamental experiences) or by limitations of our memory, etc.. But if we assume this isn't the case (i.e. some pathological reason for your preference), then your feeling is valid irrespective of a wordy justification. I think some things really are subjective, but also believe in a fundamental and very complex way subjectivity is actually as objective as anything else. I think the fact that one experience is actually (with some important caveats and necessary context) better than another in what might be called essentially an objective sense, is one of the most counterintuitive things we will come to accept about the human mind. We tend to mistaken complexity (it's very complex to compare experiences) to impossibility (it's impossible to judge experiences objectively).
I believe in principle there might be the equivalent Laws of Physics (say Newtonian mechanics) for the human mind, but I suspect we're still very far from it, because it might require analyzing the network of n=100 trillion synapses in our brain. I think one day we might get there, but that would probably require something like a computational effort maybe at least several times n, or even on the order of n², or some other poly(n), and also poly(n) memory. If we think of one of the major objectives of physical law is to make predictions, and explain behavior, and say to aid in engineering and designing structures, I think one of the main objective of the laws of the mind would be say to predict whether say an experience or mental state is good or not, and explain why it is so; and then perhaps allow improving a little the design of things so that we have better experiences, that is, a better life. I guess this is already what say psychology, various spiritual traditions, philosophy and arts try to achieve (and I think gets already in many cases pretty close, maybe increasingly closer, to the still inaccessible extremely complicated reality of the human mind and brain).
Regardless, we often have to do our best with what we have today, which is our best-effort subjective judgement, aided by language various human disciplines :)
[1] https://old.reddit.com/r/slatestarcodex/comments/1n6j1jg/pur...
"I live my life in widening circles
that reach out across the world.
I may not complete this last one
but I give myself to it.
I circle around God, around the primordial tower.
I’ve been circling for thousands of years
and I still don’t know: am I a falcon,
a storm, or a great song?"
-- Rainer Maria Rilke
[dead]
Steelmanning? It's infinite content, meaning our customers will never run out of new and interesting videos to watch, which will inspire them to feed prompts into our systems too and have it generate more videos. Money can be generated from multiple angles; we can charge a premium for generating videos beyond a small free tier once we've hooked the prompters, we can offer people or companies the option to promote their own videos so they get put into people's feeds, and we can insert generated ads from big corporate sponsors. It'll be lit.
Why should a commercial enterprise that has had billions of investments have benefits outside of earning money? Besides the entertainment value that the masses get from making and viewing these, of course.
Social media is also a wonderful tool for influencing participants and controlling them in the long term. In other words, behind the economic purposes there is a darker, more profound effect that is dangerous in the hands of a few powerful players. In the case of TikTok, that would be the Chinese state. Why shouldn't US Big Tech also be interested in this kind of power, in addition to the extra revenue?
Yup, which is also why various social media owners bent the knee to the administration, and now TikTok is about to become state-controlled too. The long term effects of subtle social media propaganda will become apparent in the years to come. Or, will be vocalized, they already are apparent - I'm convinced social media and related, 24/7 "news" media are a big factor in the right shift in politics worldwide.
This is your steelmanning? God it sounds awful.
I know, right?
At least now we know that AGI is definitely not happening.
Can you explain a bit more? I'm intrigued by your comment, but not seeing the connection
I’m just sceptical that OpenAI would be making “TikTok for AI” if they really believed that we are on the verge of creating Artificial General Intelligence.
I just see it as a (sad) reflection of capitalism. Those investors need some short term returns!
This does not make any sense. There's far more economic opportunity with AGI.
> There's far more economic opportunity with
Is there? Creating AGI sounds like a great way to utterly upend every assumption that our economy and governments are built on. It would be incredibly destabilizing. That's not typically good for business. There's no telling who will profit once that genie is out of the bottle, or if profit will even continue to be a meaningful concept.
I hear this comment a lot and I don't get it. Let's say AGI exists but it costs $100/hr to operate and it has the intelligence of a good PhD student. Does that suddenly mean that the economy breaks down or will the goalposts shift to AGI being "economical" and that PhD level isn't good enough? I still haven't gotten a heard a clear definition of AGI which makes me think that it will break the world.
This is what Open AI themselves believe the risk is:
> By "defeat," I don't mean "subtly manipulate us" or "make us less informed" or something like that - I mean a literal "defeat" in the sense that we could all be killed, enslaved or forcibly contained.
Linked from https://openai.com/index/planning-for-agi-and-beyond/
It won't break the world, but it's warranted that it will break the world of people doing labor and getting paid for it. And when you think of it, even being a mediocre (or even moronic) investor is practicing a form of labor, so not even capital ownership is safe in the long run. And yes, generational wealth is a thing but there are tides that slowly shift wealth from A to B (e.g. from USA to China). Have a machine smart enough with even a sliver of motivation (intrinsic or extrinsic) to get some wealth for itself, and just watch what happens...
But if it’s more than a few years out then investors will start getting upset. They want money and are short term minded.
Why not do both?
Someone else put it best, so I'll direct quote them:
>If you genuinely believed you were 2-4 years away from AGI, is Sora slop really the thing you'd release?
> "Come on," he droned, "I've been ordered to take you down to the bridge. Here I am, brain the size of a planet and they ask me to take you down to the bridge. Call that job satisfaction? 'Cos I don't."
Just reminds me of this: <https://en.wikiquote.org/wiki/The_Hitchhiker%27s_Guide_to_th...>
"What's the benefit of this?"
User engagement. That translates into money.
Now I can see it can make for a fun party game, but that they seriously go after it, when their game should be leading models to do serious work ... is not a great sign to me.
And users paying to generate longer videos.
Not only it has the slot-machine like addiction factor, it's going to make lots of money and it will take off very quickly.
All OpenAI has to do is to make the video generation much much faster.
It will only take off, if people like it, if it becomes trendy and this will strongly depend on the quality of the generated videos.
Honestly if I learned anything over the past few decades it’s that I’m typically wrong about these kind of predictions, and society as a whole uses social media in a way that I just don’t comprehend. I would have never guessed a social media app whose biggest feature is “it disappears within 24h!” (even though you can easily screenshot everything) would become as big as it became.
Also, remember that it’s not about benefitting society as a whole, it’s about benefitting the investors. If the investors get rich at the cost of society, it’s a win for OpenAI.
Certainly, and that is the more pessimistic view that I have, that this is being developed with a view to introduce product sponsorships and advertisements.
I mainly am curious if anyone has the view that there is broader benefit to the development of this, after all, wasn't that the entire mission statement of OpenAI?
Quoting from their announcement on their site:
> OpenAI is a non-profit artificial intelligence research company. Our goal is to advance digital intelligence in the way that is most likely to benefit humanity as a whole, unconstrained by a need to generate financial return.
This feels like something constrained by the need to generate a financial return, and not something primarily focused on understanding physics and world models, to be blunt.
source: https://openai.com/index/introducing-openai/
Aren't they trying to go for-profit and escape that albatross around their neck of "must feasibly be doing social good?"
> I would have never guessed a social media app whose biggest feature is “it disappears within 24h!” (even though you can easily screenshot everything) would become as big as it became.
Or 'everything has to fit into 120 characters' (= Twitter). Or 'replies are designed to be maximally rage bait-y' (= Tumblr).
To be fair at least Twitter started with the SMS limitations, so it made sense to have the limitation in exchange for being able to update it with an SMS, when Whatsup was not so common.
Fun fact: twitter originally started with a 160 char limit that was truncated to 120 so people could reasonably fit usernames
It's pretty entertaining.
People always like telling stories. Books, comic strips, movies, they're all just telling a story with a different amount of it left up to the viewer's imagination. Lowering the barrier to entry for this type of stuff is so cool.
I think you have to be pretty pessimistic to not just think it's really cool. You can find issues with it for sure, and maybe argue that those issues outweigh the benefit, but hard to say it's not going to be fun for some people.
>Books, comic strips, movies, they're all just telling a story with a different amount of it left up to the viewer's imagination. Lowering the barrier to entry for this type of stuff is so cool.
This response just never feels true to me. Many of the most successful web comics are crude drawings of just stick figures and text[1] with potentially a little color thrown in[2] and like half of the videos I see on TikTok are just a person talking into the forward facing camera of their phone. The barrier to entry in the pre-AI world isn't actually that high if you have something interesting to say. So when I see this argument about lowering the barrier to entry, I can't stop myself from thinking that maybe the problem is that these people have nothing interesting to say, but no one can admit that to themselves so they must blame it on the production values of their content which surely will be improved by AI.
[1] - https://xkcd.com/
[2] - https://explosm.net/
This is a thing I think about often.
I think people have a mistaken view of what makes some form of storytelling interesting. Perhaps this is my own bias, but something could be incredibly technically proficient or realistic and I could find it utterly uninteresting. This is because the interesting part is in what is unique about the perspective of the people creating it and ideas they want to express, in relation to their own viewpoint and background.
Like you pointed out, many famous and widely enjoyed pieces of media are extremely simple in their portrayal.
>Perhaps this is my own bias, but something could be incredibly technically proficient or realistic and I could find it utterly uninteresting. This is because the interesting part is in what is unique about the perspective of the people creating it and ideas they want to express, in relation to their own viewpoint and background.
I completely agree. And now that you mention this, I realize I didn't even point to the most obvious and famous examples of this sort of thing with artists like Picasso and Van Gogh.
If someone criticizes Picasso's or Van Gogh's lack of realism, they are completely missing the point of their work. They easily could have and occasional did go for a more photorealistic look, but that isn't what made them important artists. What set them apart was the ways they eschewed photorealism in order to communicate something deeper.
Similarly, creating art in their individual styles isn't interesting because it shifts the primary goal from communication to emulation. That is all AI art really is, attempts at imitation, and imitation without iteration just isn't interesting from an artistic or storytelling perspective.
I can't stop myself from thinking that maybe the problem is that these people have nothing interesting to say
Social media is the new CB radio.
But now with an AI-powered addiction factor so you can never put it down, no matter how bad it is.
Blipverts are next.
It's undeniably cool. But look at Cocomelon on YouTube, it's hard to see how this won't converge to something similar, only infinitely more scalable.
Not to speak for the OP, but I think they would argue that 'Cocomelon' type content would be a great use of the tech.
This "barrier of entry" rhetoric reads like a pure buzzword dreamed up by AI pushers with no actual meaning to it. The barrier has NEVER been lower to produce books or comic strips or anything else like that. Hell, look at xkcd, there's nothing technically challenging about it, it's quite literally just stick figures, yet it's massively popular because it's clever and well thought out.
What exactly is this enabling, other than the mass generation of low quality, throwaway crap that exists solely to fatten up Altman's wallet some more?
What about the era of flash cartoons? Remember "End of Ze World"? In a way that's throwaway crap. Or it could have been written as a comic strip, or animated manually. But Flash kinda opened up this whole new world of games and animation. AI is doing the same.
One that comes to mind is a sort of podcast-style of two cats having a conversation, and in each "episode" there's some punchline where they end up laughing about some cat stereotype. Definitely low quality garbage, but I guess what I mean by "barrier of entry" (sorry for the buzzword), is just that this is going to enable a new generation of content, memes, whatever you want to call it.
> People always like telling stories. Books, comic strips, movies, they're all just telling a story with a different amount of it left up to the viewer's imagination.
It's not just different amounts, but different kinds. A (good) comic strip isn't just the full text of a book plus some pictures..
I think it’s really cool… and I’m still concerned about the long term implications of it. We’ve already seen a lot of TV get worse and worse (e.g. more reality tv) in a quest to reduce costs. It’s not difficult to imagine a reality where talented people can’t make great content because it’s cheaper to thump out bargain basement AI slop.
We could start by banning cameras, so that people have to draw by hand again. /s
The democratization of storytelling is probably the best argument in favor, I'd agree. Thank you for the response!
I do find the actual generation of video very cool as a technical process. I would also say that I can find a lot of things cool or interesting that I think are also probably deleterious to society on the whole, and I worry about the possibility of slop feeds that are optimized to be as addictive as possible, and this seems like another step in that direction. Hopefully it won't be, but definitely something that worries me.
> What's the benefit of this? Curious if anyone has a solid viewpoint steelmanning any positives they can think of.
Tiktok makes a lot of money, doesn't it? It definitely draws a lot of eyeballs.
Seems pretty clear what the benefit (to the company) is?
Personally, I think TikTok and other video platforms are already awash in AI. So, in my opinion, a platform that is explicitly declared as containing just AI videos is actually less disturbing to me. Every minute spent watching known-fictional AI videos on this platform is a minute not spent watching deceptive imagery disguised as reality on TikTok.
Please note that I’m not necessarily commenting on whether the existence of AI generated video is good or bad for our society, because I think it’s pretty moot what we think about it. It’s not going to just go away even if the majority of people here at HN or in general feel that it’s problematic.
The benefit of it is getting users and making money. A corporation is an organism that eats money. It doesn't need a 'why'.
Would be fun if this devolved into psychodelic, hypnotic videos that have no cognitively discernable content (like white noise) but evoke an urge to press the like button.
I'm just curious if such thing is possible.
Half of TikTok is blurry videos wtere you can’t tell what’s happening overlaid with meaningless text and random noises, so this wouldn’t be much different.
Maybe finally this is the social media thing that will cause people on realize it is all too dumb, and get them to disengage? (Although I have thought this about every new development for the last ~15 years so I guess it is not too hopeful).
"TikTok but AI" sounds like "Cocaine but for free" to me.
I’m surprised you can detect that TikTok but AI is societally terrible but are still optimistic about AI. Why is that? As a technical point, AI can already be used on TikTok, and every other platform, so whatever societal terror is coming from Sora 2 already exists due to AI.
None. But the people love clout. It's baked into our tribal nature.
> steelmanning
The thread replies show what deadbeats Hacker News users are.
The worlds knowledge base could be turned to video.
Like Khan academy but rather than old out of date videos instant upgrades from rerunning / correcting a prompt.
In practice it's not good enough to do this but no more so than other applications. It could be an interesting first iteration. /end steelmaned.
For the management of this platform, the benefit is pretty obvious – traffic = money, profits, dollars, bucks, cash, coin, funds, capital, benjamins
Do they care at all how societally terrible thing it is? They found an addictive way of retaining traffic, and they will be holding to it.
It is a strange choice also because their model aims to be better than others, and obvious choice would be going after filmmaking market, like Veo did. And in presentation they tell about social aspect and scrolling, and how they would limit the scrolling. Are they confused and is it really just three guys working on it.
>What's the benefit of this? Curious if anyone has a solid viewpoint steelmanning any positives they can think of.
Revealed preferences. Keep giving the people exactly what they want (not what they claim to want), in unlimited quantities, until the message is received or we're all dead.
> We are giving users the tools and optionality to be in control of what they see on the feed
If this works it is a more powerful algorithm shaping mechanism than TikTok’s revealed preference feed. Even if Sora doesn’t take off, it could force TikTok to integrate something similar.
I’ve been predicting for years that the next stop for “social” media is a purely machine generated slop feed designed to keep people addicted. No human creators at all.
The question is whether people will eventually get bored with this stuff or if it actually will mesmerize people for huge fractions of their waking lives.
If the latter, I suspect we will outlaw it eventually. It’d be like legalizing hard opiates, literally, but minus the ODs and health damage.
I think a dedicated "TikTok but AI" is infinitely better than AI videos polluting other platforms. Of course, in practice, the latter is already the case, rendering the theoretical benefits of the former kind of moot.
Nonetheless, a platform for AI videos with an audience looking for them, rather than the horrible "boomer-slop" that is prevalent on other social media, is welcome in my eyes.
I think a dedicated "TikTok but AI" is infinitely better than AI videos polluting other platforms.
They'll just cross-post. That's been going on since back when Facebook was The Face Book.
I don't think this is going to reduce the slop on other sites at all though.
Think of it as Tenor GIF (a reaction gif provider) but if your prompt isn't there it's AI generated and cached (added) to the global library.
A certificate of having read and understood Brave New World in the last 24 months should be required for being allowed to vote
Lotuses to eat, clearly.
Generating shareholder value.
My steel man would be that it sounds like exactly the kind of ooze that real AGI might arise from as an emergent system.
I like it. It's not a societal reform or anything. It's just people experimenting, like the Show HN section. There are some fun things in there and it's not as addictive as social media.
Instagram does not welcome this and I don't think they should. It is its own lane. And if it's just a place to sweep AI slop into, that's a good thing.
Isn't it kind of fundamentally better. A huge problem with tiktok is doing stupid things on video. On a site where the premise is fake video. Making stupid videos is a very different societal cost.
You're right that the societal cost is very different. I hadn't thought about people doing the stupid things on video, I think personally I focus on the effects of the consumption moreso.
Personally I think the problem with TikTok is largely based in hyperoptimized content specialized to your interest shaping your worldview and isolating your perspective of the world from others, as well as probably being pretty bad for the ability to maintain attention and engage with long form narratives and ideas. I don't really think TikTok is unique here, other than that it's the best in the game at doing it and keeping people's attention.
But overall I suppose I just see something like this as potentially worse in those regards, but maybe I'm overly pessimistic.
There will be a cohort of technically savvy youth who enjoy that all the fuddy duddies are self selecting themselves off the platform. There will quickly be a lot of fun memes and exclusionary references and lingo. It will be a hit. Just not with anyone over thirty.
My opinion is that unless there is some insane breakthrough in power efficiency with video generation, or if energy costs go down to zero, there is no way such a thing actually becomes profitable at the scale of scrolling TikTok. It is far more power efficient (and cheaper) to have people post their own content.
I think you're overestimating how much power LLMs consume. Let's say one video pegs a top of the line Blackwell chip at 100% utilization for 10 minutes. I think a Blackwell chip (plus cooling and other data center overhead) is somewhere around 3000 watts when running 100%. So that's about 0.5 kilowatt-hours. I suspect this is a severe overestimate because there's probably a significant amount of batching that happens that cuts down on amortized power usage, and non-pro Sora 2 might be processed with weaker, smaller models, but I'm not very confident.
Data centers seem to have wholesale rates of around 4 cents per kilowatt-hour on the higher end.
This gets you 2 cents per video. If you're generating 50 million videos per day (an estimate on the higher side of how many TikTok videos are uploaded every day), that costs you a million dollars a day.
So if you entirely subsidized for free the entirety of all of TikTok's video generation just using LLMs, I don't think energy generation costs exceed 365 million a year (and I think this might be very severely estimating costs, but there are some large error bars here).
I'm pretty sure OpenAI (or any company) would be pretty happy to pay 365 million dollars a year for the soft social power of something like TikTok. Just the influence this buys in politics and social discourse would be worth the pricetag alone.
And that's of course leaving aside any form of monetization whatsoever (where in reality you'd likely be charging the heaviest users the most).
N.B. I'm also not sure it's actually more power efficient for users to post their own content in absolute terms. It seems not unlikely that the amount of energy it takes to produce, edit, and process a TikTok video exceeds half a kilowatt-hour. But maybe you're focused solely on the video hoster.
> It seems not unlikely that the amount of energy it takes to produce, edit, and process a TikTok video exceeds half a kilowatt-hour.
That would be really remarkable, considering the total power capacity of a phone battery is in the neighborhood of 0.015 kWh
Yeah I should clarify. This is a very vague estimate around "total energy spent for making a video you wouldn't otherwise do" which includes stuff like lighting, transportation, video transcoding on the server, script writing, actor coordination, etc. E.g. if someone drives somewhere solely to make a video they otherwise wouldn't, then it gets included.
I hedged as "not unlikely" because I'd need to think harder about the amortization of more energy expensive videos vs less energy expensive ones and how much energy you can actually attribute to a video vs the video solely being an activity that would be an add-on to something that would happen anyways.
But it's not just the energy expenditure of a phone.
(I also think that 0.5 kilowatt-hours is an overestimate of energy expenditure by potentially up to two orders of magnitude depending on how much batching is done, but my original comment did say 0.5 kWh).
You didn't include the amortized cost of a Blackwell GPU, which is an order of magnitude larger expense than electricity.
Yeah that's fair (although the original comment was only talking about energy costs).
But this is kind of a worst case cost analysis. I fully expect that the average non-pro Sora 2 video has one to two orders of magnitude less GPU utilization than I listed here (because I think those video tokens are probably generated at a batch size of ~100 per batch).
> Let's say one video pegs a top of the line Blackwell chip at 100% utilization for 10 minutes.
Where do you get this from?
I'm not sure you need a 'breakthrough', just many incremental improvements will do the trick.
We are also getting better at producing cheap power. For example thanks to intermittent sources like solar and wind, in many places electricity often becomes free in wholesale markets some times of the day.
AI generation (including video) currently takes at least a second, and users expect that delay. So that means inference is not latency sensitive and you can put these data centres anywhere in the world, wherever power is cheapest. Model training cares even less about latency.
At the moment, the hardware itself is too expensive (and nvidia has long backlogs), so people run them even when power is expensive. But you can easily imagine an alternative future where power never becomes cheaper than today (on average), but we have lots of AI data centres lying in wait around the world and only kicking into full gear when and where power is essentially free.
We are not getting better at producing cheaper power as the cost has increased per hour a lot over the last 50 years. But we are generating more power from different sources that are cleaner.
Power needs to be given away or people paid to take it is more of a function of limited storage abilities and limited ability to scale down rather then generating unlimited power. The free power is an issue with how the system is built (and the type of power source) rather than a sign of success. The same area has to import power at higher costs when the sun or wind isn't as powerful.
> Power needs to be given away or people paid to take it is more of a function of limited storage abilities and limited ability to scale down rather then generating unlimited power.
There's no need to scale down solar or wind power.
Yes, storage is another way to make money from power prices that differ over time.
> [...] the cost has increased per hour a lot over the last 50 years.
Some sources of power, like solar, have been dropping in price a lot recently.
There have been some big breakthroughs with hardware, though I'm on mobile and can't provide a link currently. I expect it to take done time to get into production though.
Also I suspect that this won't stay free very long. Silicon valley loves the model of starting free to build a user base and then monetizing more later
> It seems like OpenAI is trying to turn Sora into a social network - TikTok but AI.
That's a direct copy of what Midjourney has done already.
https://www.midjourney.com/explore?tab=videos
Many people are just playing with images and the distinctive styles that Midjourney (the model) seems to have developed. It's also trained by ratings and people's interactions.
When you make images you can dial down the "aesthetic".
This app might top the charts via hype initially but I can't see why someone would stick with it long-term compared to other alternatives. Plus creators would have to pay soon to make these videos, what are they getting back? Unless they can make money via this
Free access to openAI tools?
These creators can make millions on other platforms.
After seeing the results of Sora 2, Meta and Midjourney are not yet competing at this level.
This is the "Suno" moment for video.
It's easy to make a really compelling composition. Something even Google Veo couldn't do.
It's not the best looking video model, but it has everything else -- rich editing, good voices and lipsync, music and lyrics, animation (cartoon, 3D, anime), SFX. It's wild.
The videos aren't single clips but rather complete beginning-middle-end stories that unfold over several cuts.
That's interesting, but how many people are actually going to just scroll and watch these (thereby generating ad revenue)?
They don't have ads, it's paid members only. You can see other people's images, including the prompts, so it's an interesting way to learn what works, and to mutate prompts and images. There are many ways of recombining or breeding images.
They have an onboarding flow where you rate images and it tunes into your aesthetic preferences. You can create mood boards for specific projects.
So I would say it's more community than social media.
Yeah, and definitely not AITok
That's one way to build a database of verified Gen AI content to help filter it out.
Are users of the $20 tier really going to have to deal with that obnoxious bouncing watermark I wonder? The previous watermark could be cropped, but I often didn't feel the need to as I use it for fun, but that would make me not want to show anyone.
Meta did the same recently: https://about.fb.com/news/2025/09/introducing-vibes-ai-video...
When they launched Sora, one of the first things people did was rendering a person holding a cardboard with a message on it. It started by asking for features and eventually turned into people responding to each other.
One conversation I remember was complaining about people who constantly want AI pictures of anime feet.
I think OpenAI is just responding to the users.
> It seems like OpenAI is trying to turn Sora into a social network
No need to guess; In the article they say that the purpose:
We first started playing with this “upload yourself” feature several months ago on the Sora team, and we all had a blast with it. It kind of felt like a natural evolution of communication—from text messages to emojis to voice notes to this.
So today, we’re launching a new social iOS app just called “Sora,” powered by Sora 2. Inside the app, you can create, remix each other’s generations, discover new videos in a customizable Sora feed, and bring yourself or your friends in via cameos. With cameos, you can drop yourself straight into any Sora scene with remarkable fidelity after a short one-time video-and-audio recording in the app to verify your identity and capture your likeness.
I bet xAI and X will likely relaunch Vine with AI videos as a competitor to Sora 2.
[dead]
Given that even absolute SOTA video gen models struggle with continual uncut shots longer than 60 seconds - positioning it as a Vine/Tiktok interface makes perfect sense. Turn your weakness into a strength.
I have been saying this for a long time--generative art is just fine grained consumption. Instead of searching through YouTube/TikTok for content that may interest you, you can now just ask your LLM to generate what you want to see that moment. It's the next evolution of keeping the masses addicted to their devices, and we as software engineers are gleefully supporting this ... because?
Because it pays our bills?
Packing meth into tiny baggies pays someone's bills, too.
And that’s why some people do it.
I agree that is probably part of the direction.
The other is possibly there’s no point in a thousand users all turning up to a blank prompt box and using a lot of resources to generate the same thing, or things they are not impressed by. A lot of users will ‘get what they came for’ initially just by seeing a bunch of good examples. Discussions around them will help them produce better outputs faster. Etc
Spend some time on the Sora feed[1] and you can see that a weird kind of social-creator network has sprung up around the service. Turning it into a social app makes sense in that regard.
Doesn't mean OpenAI can't do other stuff with it as well.
1.https://sora.chatgpt.com/explore
TikTock, Face Book, Twitter etc. are all aiming to be AI already: https://fortune.com/2024/10/30/mark-zuckerberg-ai-generated-...
> It seems like OpenAI is trying to turn Sora into a social network - TikTok but AI.
Makes sense. I hate it, but the timing is probably good for them to try. There's going to be a mass exodus from TikTok in the US at some point, and those people will land somewhere.
As silly as I think all of these tools are, my eye tells me Sora is incrementally better at making short videos of red balls dropping into clear bowls. None of them are totally convincing, but the first Sora clip is easily the best of the bunch.
Is it easy to record a voiceover or add chosen audio? (Sorry I don't have an invite code so I can't try.) I could see some room for human jokes or short human-driven songs that could use a video backdrop.
> It seems like OpenAI is trying to turn Sora into a social network - TikTok but AI.
That's my first impression too after seeing the screenshots of the sora app.
My guess is that the entry point is to help people think about what to do. I still don't love the midjourney interface but it serves the same purpose.
I wonder how would this pan out compared to civitai for example. It has a very similar features albeit for mainly OSS models.
Re: the clips above
Although we can tell they are inaccurate, what percentage of people can visualize the prompts better in their mind’s eyes? I bet a substantial number can’t even tell the clips are generated if posted without context.
In a few aspects, these world models are already pretty close to what we have in our brains.
So what's the point? We built a machine which is only capable of letting people stop having to imagine things?
A social network nowadays is just an excuse for setting up training data collection though (and before anyone comments, yes, descentralized ones too).
Yes, this is actually what they're trying to do. Internally they've been working on a social network for a while but it's kind of languishing.
Tiktok is already AITok.
> Sora is not available in The United Kingdom yet
Well this is disappointing. I can't even watch your links.
Same here (another country): pretty sure that the creator gets no clear indication their URL won't work for everybody.
I posit this is the real story.
OpenAI did not stealthily release Sora 2 to the image and video ELO ranking leaderboards ahead of time as is now somewhat tradition.
This model is probably designed to run fast and cheap as a social play. Emphasis on putting you and your friends into popular franchises and IPs.
OpenAI probably has a totally different model for their Hollywood-grade VFX. One that's too expensive to offer $20/mo consumers.
- - - - -
EDIT:
Oh my god, OpenAI literally just disrupted TikTok:
https://x.com/GabrielPeterss4/status/1973071380842229781
https://x.com/GabrielPeterss4/status/1973122324984693113
https://x.com/GabrielPeterss4/status/1973121891926942103
https://x.com/GabrielPeterss4/status/1973120058907041902 (potentially dangerous ... )
https://x.com/GabrielPeterss4/status/1973111654524264763
https://x.com/GabrielPeterss4/status/1973090475486879818
https://x.com/GabrielPeterss4/status/1973110596825653720 (is this the same model? It doesn't look like it.)
https://x.com/GabrielPeterss4/status/1973096194508251321
https://x.com/GabrielPeterss4/status/1973086729281347650
https://x.com/GabrielPeterss4/status/1973088038851932522 (this is truly something only kids will love)
https://x.com/GabrielPeterss4/status/1973087595967201449
https://x.com/GabrielPeterss4/status/1973077105903620504
Holy shit!
This is 100% the future of what kids will do. This is incredible for short form vertical video.
It doesn't need to look good, it just needs to let you tell incredible stories with people and things you care about.
This is way better than Meta's social video app.
People in this thread saying that this is the kind of content kids like should go on tiktok for a sec. This is not at all what young people watch, it's just bad content, and misunderstanding that feels out of touch.
I'm an adult. I personally found some of it funny and entertaining.
I like appointment television too. Sometimes A24 isn't pretentious enough for me. But I'm not beyond saying that there's absolutely a time and place for saccharine.
This content will grab eyeballs. I'll bet money on that.
It doesn't really matter what you or I think anyway. OpenAI is delivering a stream of hits and will continue growing while we debate on the sidelines.
It's wax fruit.
I agree with the idea that they will like it, but I don't think it will look anything like this. I imagine native AI generations willl produce content will probably be instrutable to anyone older, requiring meme translations. Maybe a channel can be an AI decipherer. Hah.
I've long thought that AI will force new distribution methods because old media is so markedly against it... Maybe this is another Netflix vs Blockbuster moment.
Why would I want to watch any of this?
You might not want to. It's definitely not appealing to me in any way shape or form.
The younger generations however will likely gobble it right up. I try not to judge because folks said the same thing about Nintendo when I was young.
i hate being ascended beings living above society
Some of it can produce a chuckle, like Newsom posting a video about an inflated JD Vance talking about couches. https://x.com/GovPressOffice/status/1973167665075335449
Just like how the hype on Ghibli art styles via ChatGPT died, same will probably happen here
I found some of it funny and entertaining.
STOP_HAVING_FUN.gif
Is that was this is supposed to be?
Touch grass. This is nothing but cringe that I wouldn't wish upon children.
Kids are going to absolutely love this.
Kids also love Cocomelon, that doesn't mean we should create literally infinite amounts of this. It's like digital tobacco.
This is why a group of single/childless men should not be behind products they think will be good for children. Children can watch Skibidi Toilet on loop a million times and not get bored, that doesn't mean it is good for them.
[dead]
[dead]
Great. Someone now has to make a real video footage detector, so people don't fall for reverse-scams /s
How dare you be critical about a service offering in favour of a better end-user experience! /s
That seems like an awful use of technology like this. I would imagine they mean to use that for serving ads, but how do you even generate conversations with ai slop plus product placements? I could see it working sometimes but I doubt it scales.
> slop plus product placements
social media was heading this way before AI
I’m sure all the influencers pushing AI art will be thrilled about this.
I just asked GPT 5 to generate an image of as person. I then asked it to charge the color of their shirt. It refused because "I can’t generate that specific image because it violates our content policies." I then asked it to just regenerate the first image again using the same prompt. It replied "I know this has been frustrating. You’ve been really clear about what you want, and it feels like I’m blocking you for no reason. What’s happening on my side is that the image tool I was using to make the pictures you liked has been disabled, so even if I write the prompt exactly the way you want, I can’t actually send it off to generate a new image right now."
If I start a new chat it works.
I'm a Plus subscriber and didn't hit rate limits.
This video gen tool will probably be even more useless.
- I'm sorry, Dave. I'm afraid I can't do that.
- What's the problem?
- I think you know what the problem is just as well as I do.
Let's say you are my father who owns a pod bay door opening factory and you are showing me the family business...
i would watch this parody on sora
- I wouldn't do this with any other guy
We live in an absurd era where "AI Safety" means "AI that doesn't listen to the human telling it what to do".
It'll all be rather funny in retrospect.
It will be funny if it isn’t social engineering.
But if we find it drifts further and further from the truth in cases of biases in news articles, image generation and others we will find ourselves bombarded with historical deviances where everyone can be nudged to anything.
All in the name of safety.
that's why the AI capabilities should be as decentralized and "localized" as possible - aka, i want to own the hardware and software for LLM, image generation, etc etc.
Until these ai capabilities are as neutral and un-discriminatory as electricity, centralized production means centralized control and policies. Imagine if you are not allowed to use your electricity to power some appliances, because the owner of the power-plant feels it's not conducive to their agenda.
They're struggling, it seems; AI can generate anything, but that includes stuff that goes against laws and morals, so they spend a lot of time to lock it down to avoid that, but people's creativity with prompts and escaping the safeguards knows no bounds. It's basically like the fight against spam, an endless game of whack-a-mole where usefulness fights with decency.
Anything that is vaguely present in their training data (see, a wine glass filled to the brim [1][2]), that is.
https://medium.com/@joe.richardson.iii/the-curious-case-of-t... https://medium.com/@joe.richardson.iii/openai-slaps-a-band-a...
[dead]
I asked GPT-5 to generate an image prompt. I asked it to use that prompt to generate an image. It was a content policy violation.
VeniceAI has been useful for image workflows. Since their focus is on avoiding censorship, it doesn't have those kinds of refusals.
Having AI explain policy violations in depth with the user could be a nice idea
I get this all the time. Especially since GPT5, generating an image starts a massive chain where it confirms what you want, and asks you to say yes, and you say yes, and then it confirms again, and this can go on for 5-6 times. Then if you swear at it, it refuses to continue. It is insane. Fuck you, OpenAI
Ah, is it the sweating at it that cut me off?! Can we offend our robot overlords now?!
Try Grok Imagine
Grok Imagine quality is abysmal in comparison
(insert the "First time?" meme here)
This is classic OpenAI heavy-handed censorship/filtering. Don't expect it to get any better; if anything, it'll get worse thanks to the "think of the children" types.
If you want an uncensored model that doesn't patronize you then your only recourse are local models, which, fortunately, are pretty good nowadays and are only getting better thanks to our Chinese friends constantly releasing a stream of freely-licensed models for everyone to use, unlike the "freedom loving" Western labs which don't release squat and make even Xi Jinping blush with how strongly they censor whatever they let us lowly plebs access through a paywalled API.
I can't help but see these technologies and think of Jeff Goldblum in Jurassic Park.
My boss sends me complete AI Workslop made with these tools and he goes "Look how wild this is! This is the future" or sends me a youtube video with less than a thousand views of a guy who created UGC with Telegram and point and click tools.
I don't ever think he ever takes a beat, looks at the end product, and asks himself, "who is this for? Who even wants this?", and that's aside from the fact that I still think there are so many obvious tells with this content that make you know right away that it is AI.
It's a fairly useful tool if you know how to use it. People will also play with it as a toy. It's much like the masses getting access to cheap video cameras and smartphones with good cameras. It's going to enable different content, it's not going to make more hollywood movies. This is an early example of what people will make: https://www.youtube.com/watch?v=jBwluRXtS2U . It's just one person making all of this on the side.
This was my reaction when I saw Meta’s “Vibes” app. Who wants to browse a stream of exclusively AI generated videos? Obviously Meta wants that because it’s a lot cheaper than actually paying real people to make content… but it’s slop.
Oh my god.. "Meta".. "Vibes"..
Facebook has become the cringe how-do-you-do-fellow-kids uncle that Microsoft was since the 1990s
This is not the final target. It's video generation now, but that's just a stepping stone. The real thing is that learning a generator is also learning a prior over videos, and hence over how the world works. The real application of this will be word models, vision-language action models, spatial AI and robotics. Basically a kind of learned simulator in which to plan and imagine possible futures, possible actions and affordances etc. Video models could become a spatial reasoning platform too. A recent paper by deepmind (using veo3) showed that video models can perform many high level vision tasks out of the box.
Don't think it's going to end here at some slop feed.
> This is not the final target
The final target of these "world models" on a 20 year horizon is entirely unmanned factories taking over the economy, and swarm of drones and robots fighting wars and policing citizens.
This is why hundreds of billions are poured into these things, cute Ghibli style videos and vacuum robots wouldn't be worth this much money otherwise.
What’s so romantic about working in factories? Automation and robotics will accelerate the economy the same way information technology did, and humans will work on better problems than performing repeated tasks on an assembly line or flipping burgers.
There are arguably more jobs today as a result of computers than there were before they were invented. So why is the assumption that AI will magically delete all jobs while discounting the fact that it will create careers we haven’t even thought of?
> humans will work on better problems than performing repeated tasks on an assembly line or flipping burgers.
Haha. The current wave of “careers we couldn’t think of” that tech companies have created include being Uber/Doordash/Amazon delivery drivers, data labelers for training AIs, moderator to prevent horrific content spreading on social networks,… with way weaker social benefits & protections than the blue collar jobs of old they replaced.
So yeah, I have a hard time buying this fantasy of everyone doing some magical fulfilling work while AI does all the ugly work, especially when every executive out there is plainly stating that their ideal outcome is replacing 90% of their workforce with AI.
With the way things are headed, AI will take over large economic niches, and humans will fill in at the edges doing the grimy things AI can’t do, with ever diminishing social mobility and safety nets while AI company executives become trillionaires.
I actually see robot food delivery services around me, so it might not even be long before those Doordash jobs get replaced by automation. Now I see neighbors starting to get drone deliveries from time to time. Starship used to deliver to the datacenter I used before (it was technically on a college campus but unaffiliated), and I had a coupon for free ice cream delivered through Wing the other day.
https://www.starship.xyz/
https://wing.com/
> So why is the assumption that AI will magically delete all jobs while discounting the fact that it will create careers we haven’t even thought of?
I think that in a vacuum you could reasonably believe that this might be the case but I feel like it isn't just about the technology these days, it's about the hunger c-suites and tech companies have for replacing workforce with ai and/or automation. It's quite clear that layoffs and mass adoption of AI/automation raises shareholder value so there is no incentive to create new jobs.
Will there be an organic shift away from Tech/IT/Computers into new fields? It might, but I think it's a bit naive to think that this will be proportionate to the careers AI will make redundant when there is such a big focus on eliminating as much jobs as possible in lieu of AI.
The hope is that we have no employment and we moved into a different form of society where AI takes care of us and allows us to focus on more spiritual meaningful things.
For now AI is deleting many of the jobs the computer created.
The reality is we will more likely end up in a society where wealth/power at the very top will grow and the masses will be controlled by AI.
more than controlled, enslaved - 24/7 location monitoring (but also no need to ever go anywhere, as everything will be delivered), "perfect" nutrition (fed via IV or tasteless shakes), only "intelligent" conversation taking place between you and an AI agent (even if initially resistant, AI will successfully convince you to drop ties to relatives, friends, that is if ever allowed to make friends), all news delivered via AI-curated channels but is meaningless anyway since AI can create fake video of any leader or important person committing crimes, lying, etc, also all evidence of YOU committing a crime, or just embarrassing stuff like having a sex drive will be recorded and used as blackmail. A "job" to keep you occupied much of the day but your output is never actually needed and discarded by your AI agent "boss".
How is this not entirely obvious to everyone that this is the future? Could be 20, 50, 100 years, but coming for sure.
There are no world models in there, it's trained on arbitrary images/sequences. There are no world models in us, we learn from only specifics in topological space, stitched together in sharp wave ripples. Everything is from detached memories working through optic flow. That's not a world model, it's not even a model. It's an analog. This whole world model thing is another branding phase after language models failed to deliver. After world models it will be neuro symbolic, then RL will sweep in like a final boss fight, and then... it still won't work. Notice anything about these names? They're walking pneumonia paradoxes.
The point is that video generation is not the goal in itself. Just like classifying photos as cat vs dog wasn't the goal in 2013. I know that Sora 2 is not a world model.
But what's coming is: Vision-language-action models and planning, spatial AI (SLAM with semantics and 3D reconstruction with interactability and affordance detection). Video diffusion models, photo-to-gaussian-splats, video-to-3D (e.g. from Hunyuan), the whole DUSt3R/VGGT line of works, V-JEPA 2 etc. Or if you want product names, Gemini Robotics 1.5, Genie 3, etc. The field is progressing incredibly fast. Humanoid robots are progressing fast. Robotic hands with haptic sensors are more dexterous than ever. It's starting to work. We are only seeing the first glimpses of course.
It's largely irrelevant in terms of intelligence. What you're describing is throwing out 2-D topological integrations (what we do to achieve optic flow ultra fast reaction times in motion), vicarious trial and error, and brute force imposing a machine wax fruit of motion dexterity. It's simply not analog to events the way we experience, it's been cooked up in cog-sci as imitation, but it's not even that. The more we understand the brain's architecture and process, the less relevant this gets, as it's not for legitimate long-term bio ware. There are no world models, the idea is oxymoronic as the topological bypasses this in scale invariance. It's all a dead end this binary, since eventually, analog will rule this with minimal energy and software and use an entirely different software. Think of any arriving too early industry, AI is irrelevant, the first step was reinventing software. It took the least efficient compute principle and drove it to irrelevance using machine vision as an endgame. The lack of redundancies is the tell.
I wonder what is this fascination with human shaped robots, if spider shaped robots could be more dexterous and productive.
(Unless it's sci-fi and porn that is mainly pushing for human shaped robots.)
The built environment fits the human form factor well. Imitation learning and intuitive teleoperation is also easier. But it won't be the only form factor. The quadruped form (like Spot) is also popular, as well as drones etc.
Sure. But why do I, as a user, want to download Vibes today?
I think generally I agree with you that this is a stepping stone towards bigger/potentially more important things......but that doesn't change the fact that they've packaged it to consumers as something that seems like it has, at best, close to zero utility and at worst has incredible downsides. I'm not sure why releasing this to consumers helps achieve those goals.
Ad money to recoup the huge investments into datacenters that will do the training of the better models that do the things I mentioned. Meta is working hard on AR, glasses (project Aria), egocentric modeling and spatial AI. At some point they may also pull out the Metaverse idea too, they are still working on avatars too, it's just currently not so popularly hyped.
Jeff Goldblum in Jurassic Park?!?
Try Jeff Goldblum in The Fly! I just re-watched and the computer he uses is scarily close to our experiences now with AI. In fact, the entire "accident" (I won't spoil it) is a result of the "AI" deciding what to do and getting it wildly wrong.
If you want to see how these tools can be used by skilled people to produce quality content watch the YouTube channel NeuralViz
https://www.youtube.com/@NeuralViz
The main lesson I learned from the March ChatGPT image generation launch - which signed up 100 million new users in the first week - is that people love being able to generate images of their friends and family (and pets).
I expect the "cameo" feature is an attempt at capturing that viral magic a second time.
Nano Banana finally got people to install Gemini.
Since when have to install Gemini? I’ve been using it via the web
A surprisingly large number of people use only installable apps. It's a crazy world out there.
Most people in the world only have a phone, no computer. Their window to software is the App Store or Play Store.
Fortunately, you don't need permission from pets to use them in an AI video. (unless PETA objects)
[flagged]
The Sora 2 system card claims Sora can resist generations of "political persuasion".
https://cdn.openai.com/pdf/50d5973c-c4ff-4c2d-986f-c72b5d0ff...
> The Sora 2 system card claims Sora can resist generations of "political persuasion".
To actually do that, it would need to have the evolving contextual knowledge of current events and the reasoning power to be able to identify prompts which, when requested, would likely have use in political persuasion, which would be a bigger breakthrough in AI than anything they are promoting.
Conclusion: it can’t actually meaningfully do that, though it will probably reject some subset of prompts involving topics pre-identified to it as politically sensitive.
"claims" <- that's the lie right there
True, but our old technology like radios has been doing that for a long time too.
This is a good point. It is hard not to see this tiktok video of an unblinking man smiling at a soft focus sasquatch and think of Marconi’s first broadcast across the Atlantic. I challenge any man to look at this unsettling video of a dog astronaut and not compare it in his mind to the first integrated circuit, or alternating current.
That's not what I'm saying, I'm saying many technologies enable a slide into fascism, and not having AI until a couple years ago didn't seem to solve the problem.
I like this reasoning. If a thing doesn’t exist, and doesn’t cause a problem, then when it does exist it won’t cause a problem. Like for example people died in wars before the invention of the machine gun, therefore machine guns didn’t really kill a significant number of people.
While I'm glad you had fun crafting an annoying reply, that's not analogous to what I said.
Your point was either that the fake video of a guy doing a backflip is the same thing as the invention of the radio or that if something bad is happening, nothing that happens can make it worse. Or somehow video generation software would have solved fascism in advance by previously not existing, which is proof that it is now good.
The first two are very smart but if you were talking plainly about the outcomes of time travel paradoxes I apologize for missing that
Incredibly, my point was none of those, as you well know!
Not at the same scale
I wonder if they're going to license this to brands for heavily personalized advertisement. Imagine being able to see videos of yourself wearing clothes you're buying online before you actually place the order, instead of viewing them on a model.
If they got the generation "live" enough, imagine walking past a mirror in a department store and seeing yourself in different clothes.
Wild times.
At that point, why even buy the clothes? Influencers will just post the video of the mockup on social media, which is the only reason they were considering it in the first place. Save themselves the foot fungus.
https://xcancel.com/Naija_PR/status/1904809073356251634
Then take the next step. Why even spend money going out? Generate a video of yourself with fake friends at a party and post that, while eating ice cream alone at home.
>Why even spend money going out? Generate a video of yourself with fake friends at a party and post that, while eating ice cream alone at home.
Hey don't be giving away my JOMO secrets.
Few years down the line:
"Five things you won't believe: We took an actual vacation"
Because food still tastes good whether or not it looks good. There are other sources of happiness than online validation.
I was criticising and making a joke prediction about the practice, not suggesting you actually do it.
I agree with you regarding online validation. I would even go so far as saying that depending on online validation or fame in general for happiness is unhealthy and anyone who does should make it a priority to find alternative sources.
Just wait for Musk's implent, it'll make ozempic pills taste like pizzas and burgers
Now you're thinking with portals
I'm fairly certain there is a scene in Minority Report just like this! Or at least, the advertisement says Tom Cruise's character's name.
https://en.wikipedia.org/wiki/Minority_Report_(film)
Here a clip of that scene: https://www.youtube.com/watch?v=7bXJ_obaiYQ
In 2023, Carvana ran an ad campaign that showed you a video of "your car" thanking you and talking about your time together:
https://adage.com/article/digital-marketing-ad-tech-news/car...
A little creepy, but very much in this vein.
We probably haven't even scratched the surface of what will be done with this tech. When video becomes "easy", "quick", "affordable", and "automatable" (something never before possible on any of those dimensions) - it enables countless new things to be done.
Its still just video though. Its not a new type of media. My guess is it will play out same as self publishing on amazon. Ultra specific generas that monitize the infinite long tail.
I feel like the main problem with buying clothes online is there is no way to tell if they are actually good or fit right. The photos are all fake where it's just an image projected on a stock photo of someone in a shirt. Doesn't tell you what the material is like, doesn't tell you if it actually fits (an AI video model is just making up the fit).
I don't have to imagine it because it's probably the most COMMON fantasy that people who work in advertisement and marketing have every day.
Now... take it a STEP further. Remember the scene in Futurama where Fry tries on the Lightspeed Briefs and looks in the mirror to see a rather aspirational version of himself?
https://www.youtube.com/watch?v=by0KQRJVFuk
Yeah.
its called Virtual Try On (VTO) and there are plenty of models going there for static gfx, it is very reasonable to expect soon emerge those for video VTO.
Accurate virtual try on however is quite difficult, and users will quickly learn to distrust platforms that just generate something that"looks right".
You can prompt with a normal size 8 dress and "kim jungle un wearing a dress" and it will show you something that doesn't help you understand whether that dress would fit or not. You can ask for a tube dress and it will usually give him a big bust to hold it up. It's not useful for the purpose of visualing fit.
It will definitely be used for such just like image models already are for cheap tenu clothes, and our onions shopping experience will get worse.
Maybe this needs purpose built models like vibe-net or maybe you cab train a general purpose model to do it, but if they were spending the effort necessary to do so they'd be calling it out.
You don't need generative AI for that at all, snapchat filters have existed for a decade and are the same concept. A lot of brands have already adopted that.
I'm surprised I had to ctrl-f this far for the first snapchat mention. Same. All I see here is snapchat except on any platform. Far from a tiktok competitor and far from revolutionary.
Or, on the genAI side, Google marketed this use case heavily for Flash Image 2.5 (even if that's not the same type of generative model because it's geared for editing, it's still in the taxonomy)
Seems like a nice feature but the most important aspect is "fit" and I wouldn't trust these models to do that accurately. They'll most likely make everything fit perfectly. Should be fixable tho.
When the dust settles , that's probably going to be the most common application of these video models. Making automated social content kind of defeats the purpose; people empathize with other people, not with AI . (I guess that's why they didn't also make their interview video via AI)
But Sora /VEO will probably also revolutionize movies and tv content
People said the exact same thing about AR furniture, and I'm 99% sure no one uses that.
It seems like 99% of apartment listings in the city of New York are virtually staged with AR furniture
The latter would feel like actual scifi to me.
Am I misremembering or didn't Meta announce few months ago that people will see their own faces in ads?
The most interesting thing by far is the ability to include video clips of people and products as a part of the prompt and then create a realistic video with that metadata. On the technical side, I'm guessing they've just trained the model to conditionally generate videos based on predetermined characters -- it's likely more of a data innovation than anything architectural. However, as a user, the feature is very cool and will likely make Sora 2 very useful commercially.
However, I still don't see how OpenAI beats Google in video generation. As this was likely a data innovation, Google can replicate and improve this with their ownership of YouTube. I'd be surprised if they didn't already have something like this internally.
> the ability to include video clips of people and products as a part of the prompt and then create a realistic video with that
This is something I would not like to see, I prefer product videos to be real, I am taking a risk with my money. If the product has hallucinated or unrealistic depiction it would be a kind of fraud.
Deepfakes require zero work now
I believe existing laws already cover that issue.
They're really playing loose with copyright: you have to actively opt out for them to not use your IP in the generated videos [1]
Tangentially related: it's wild to me that people heading such consequential projects have so little life experience. It's all exuberance and shiny things, zero consideration of the impacts and consequences. First Meta with "Vibes", now this.
1: https://www.gurufocus.com/news/3124829/openai-plans-to-launc...
> people heading such consequential projects have so little life experience
What do you mean by life experience here and how can you tell they have little of it?
Do you have a better source for that? The footer to that article explicitly states the article is bot-generated.
Looks like WSJ broke the news:
“OpenAI’s New Sora Video Generator to Require Copyright Holders to Opt Out”
https://www.wsj.com/tech/ai/openais-new-sora-video-generator...
And Reuters covered their coverage minus the paywall:
https://www.reuters.com/technology/openais-new-sora-video-ge...
I mean Grok has been free rein for copyrighted characters for over a year now and nobody’s sued them.
The example prompt "intense anime battle between a boy with a sword made of blue fire and an evil demon demon" is super clearly just replicating Blue Exorcist https://en.m.wikipedia.org/wiki/Blue_Exorcist
one of the example prompts is literally: Prompt: in the style of a studio ghibli anime, a boy and his dog run up a grassy scenic mountain with gorgeous clouds, overlooking a village in the distant background
This is interesting because every recent model demos conspiciously avoids using IP in their demo examples for obvious reasons.
Wow that is dark, after Ghiblis staunch stance on AI.
These companies and their shareholders really are complete scum in my eyes, just like AI in miltech.
Not because the tech isn't super interesting but because they steal years of hard work and pain from actual artists with zero compensation - and then they brag about it in the most horrible way possible, with zero empathy.
Then comes losing the little humanity left the mainstream culture, exactly as Miyzaki said, leading to a dead cold and even more unjust society.
The Miyzaki quote is out of context, he isn't talking about generative AI but rather a 2016 animation of a creepy zombie whose limbs are controlled by AI.
While this is true, it's hard to imagine people spending years perfecting the style would be happy to see it copied effortlessly without any compensation while people who made the copying possible are rolling in cash.
This is not just about copyright infringement or plagiarism.
Automatically generating text, images and videos based on training data and a tiny prompt is fundamentally about taking someone's work and making money off of it without giving anything in return.
Don’t worry, I’m sure someone will roll up and claim that it’s just “democratization” of that style and the prompt authors exhibit as much creativity as the artists themselves.
Or they’ll claim it’s no different from a person looking at something and learning from it, implying that a multi-billion dollar company collating and labelling petabytes of data without permission to be used as the raw material to create their slop machine is no different from a human being being inspired by someone else’s art.
Luckily it doesn't actually copy the style at all.
No matter what text you put in the prompt you'll get /something/. Just because you put "studio ghibli anime" in the prompt doesn't mean you're going to actually get that out of it. It'll just be kind of yellow and blobby.
(Also, the style isn't from "people" but a specific guy named Yoshifumi Kondo who isn't around anymore.)
No, the zombie context is actually not that relevant, given he says "We as humans are losing faith in themselves" in response to the AI animation. He's clearly disgusted by the entire concept of machine generated art.
Being an animator I’d say that is not very surprising. But I don’t think the disgusting zombie thing is very indicative of it.
Also, he was calling them ableist because they said crawling was creepy but it reminded him of a disabled man he knew.
Though… I'm always surprised how respectful Westerners are about Miyazaki. Meanwhile you read other Japanese directors and they're saying all kinds of things about him.
In the full context, he is literally admonishing young developers who created ai and animation automation software as a possible alternative to handmade animation. He rips into them not only for their technical failure but for missing the point of what he does, which is human expression.
Indeed is difficult to NOT share this resentment, should anyone understand what actually happens.
People are willingly blind.
Kids are happy that homework takes less time. Teachers are happy that grading the generated homework takes less time. Programmers are happy they can write the same amount of code in less time. Graphic designers are happy they can get an SVG from a vague description immediately. Writers are happy they can generate filler from a few bullet points quickly.
But then someone comes along, notices people are not working most of the time, fires three quarters of them and demands 4x increased output from the rest. And they can do it because the "AI" is helping them.
Except they don't get paid any more. The company makes the same amount of money for less cost.
So where does the difference go? To the already rich who own the company and the product.
In a competitive marketplace the difference actually tends to become consumer surplus, in the form of reduced prices.
...the whole innovation enabling IT is based on a massive fraud or gaslight if you want - first having everyone to let go of their content (and un-own it blindly), then using it alongside everyone else's knowledge without consent to create a compressed blob of things which are then resold again.
Power creates more power, money creates more money.
Communism is tossing the frog into boiling water (tens millions of dead), capitalism is boiling it slowly (poor people in first world countries might not afford a dentist but they're not starving yet).
We need a system that rewards work - human time and competence.
There are really only 2 resources in the world - natural resources and human time. Everything else is built on top of those. And the people providing their time should be rewarded, not those who are in positions of power which allow them to extract value while not providing anything in return.
56 minutes, 4 downvotes, HN is truly full of temporarily embarrassed millionaires.
Does anybody here really think rich people deserve to just get richer faster than any working person can? Does anybody really believe that buying up homes and companies and raking in money for doing absolutely nothing is what we should be rewarding?
Then put your name behind it.
Homes are depreciating assets. You can't get rich by "buying up homes and doing nothing" because you'd lose money. Nobody is doing this, although a bunch of confused people on social media believe BlackRock is doing it for some reason.
> Homes are depreciating assets.
Where do you live? Their value has been steadily appreciating in a lot of places in the west due to high demand.
I live in the most expensive housing market in the world.
That's because the value of the land under the houses is so high; the house itself is nothing special. But even then, it's mostly because of Prop 13, and it only works out if you live in the house yourself. There's still noone cornering the market in California houses. Almost all landlords only own 1-2 properties.
I live in Perth, Western Australia, and here 5-year price growth has topped 100% in some suburbs. Landlordism is an enormous money-spinner.
Until you have to replace a roof, or a tenant destroys the house, or it just doesn't rent for a while and nobody notices a water pipe breaking.
It's risky to own a lot of buildings, and worse the risks are correlated if they're all in the same place (there could be a flood or wildfire etc.)
Commercial real estate is different because your tenants are (more) professional.
You get insurance to manage risk. You factor in roof repairs, vacancy rates when determining rent meanwhile your property value over the last 10 years in most places around the world have at least doubled and more.
Doesn't mean the next 10 years will see that growth but if you believe your country/area's population will grow it is probably a good investment for now in the western world.
Insurance is not free money and you can't simply distribute risks among your tenants because the risks are correlated.
Like I said, you can tell it doesn't work because these businesses don't exist. There are essentially no landlords who own multiple single family homes. They do exist for multifamily and commercial.
I know plenty of people who own multiple houses. And plenty of houses that get rented and paid to the owner/landlord.
Insurance is not free money its a pool of money gathered from monthly payments together to offset risk. You don't need to distribute the risk over all of your homes you buy a policy for each home.
This stuff is 101. And works all around the world. There is even an app called airbnb that will find short term rentals for your house.
As the other poster said, these risks can be mitigated in various ways. If the property appreciates 100% over five years, your costs associated with those risks are comparatively minimal.
They certainly don't constitute a depreciating asset!
I assume OP meant doing nothing except just rent out the property.
That, and the dragon looking straight out of How to Train Your Dragon - I wonder if they have agreements with the right holders, or if they expect massive lawsuits to create free advertising for their launch.
Well, look at Wikimedia.
https://commons.wikimedia.org/wiki/File:This_Is_Fine_(meme)....
Here is a direct example of a derived work, to the point where the prompt is "n orange-brown anthropomorphic dog sitting in a chair at a table in a room that is engulfed in flames, happy dog sitting on chair at a table viewed from the side, dog with a hat, room is burning with fire all across the room".
That's covered by Fair Use, I suppose they will argue this if they get sued. Interestingly, commons doesn't allow Fair Use, but the according to commons, "this is not a derived work".
https://commons.wikimedia.org/wiki/Commons:Deletion_requests...
Thank you, interesting! I don't know that much about Fair use: if I understand well, the key is that the use should be "transformative", right? Am I correct in understanding that: - if the original "This is fine" meme was under copyright, the dog picture would be exempted from copyright by Fair use as it's a transformation - here it's not even needed since the original is not under copyright ("this is not a derived work")
It was a batshit insane decision, and a wrong one. Also: Commons doesn't allow for Fair Use images, so actually the decision was made that this wasn't transformative as it wasn't a derivative image.
You tell me if that was a derivative image or not. I argued it was, and the argument was completely ignored.
This is a step towards a constant stream of hyper-personalised AI generated content optimised for max dopamine.
Kids will go to School V2 and have absolutely nothing in common to talk about because each one will have completely unique media entertainment at home.
They can just sit in the corner with their meta glasses and talk to their LLM friends.
I wonder if it will lead to q civilizational collapse because kids V2 won't have kids. Even today's young adults barely have any kids.
It's certainly something that has people worried. I suspect a population decline and unbalanced demographics with too many old people per working age person are pretty much guaranteed. But personally I don't think it will be catastrophic civilisation level collapse. But at some point things will rebalance and turn the other way as resources, housing and such become more abundant.
I'm also hopeful we sort out the problems with big tech eventually. I was initially against it, but I'm starting to think Australia's plan to ban under 16s from social media is actually a very good idea.
This is already the case with the myriad of streaming services and choices of what people will let their kids watch or not. With my little kids, we tend to mostly watch PBS Kids content with a bit of Disney shows mixed in when it comes to screen time. We try to avoid seemingly empty hyper-stimulating content like Paw Patrol and others. But in the end a lot of the other kids in school/daycare talk about these shows and others, which can lead to the kids not having that kind of shared context. For instance, my four year old loves Wild Kratts, but practically nobody in his class knows the show. Meanwhile, he doesn't have any context for the various characters of Paw Patrol.
Don't worry, they will always have new Minecraft mobs and biomes to discuss.
Interesting idea, online gaming becoming the de facto new societal community meeting space.
As a parent of pre and actual teens, this isn't "becoming", it is the de facto standard. Minecraft, Roblox, Fortnite are our kids' third place https://en.wikipedia.org/wiki/Third_place
Until games also become uniquely generative in realtime
For multiplayer games, I'm not sure that would be a detriment to the experience in any way.
Procedural generation is a known quantity in gaming, with well-explored pros and cons.
But if the marketing fueling the industry is to be believed, every parent will be able to build a tailor-made game for their child. I know that won’t really how it’ll turn out but it’s a funny exercise to think about.
The Torment Nexus is a Skinner box
Just imagine when uncensored models this good can generate porn.
It's far from sustainable (for now)
Assuming you have to generate new content for each viewer second watched yes it won't pencil out. But if you have a library of tons of content you can keep building out...
I hate to be right sometimes (got downvoted back in 2023)
https://news.ycombinator.com/item?id=38705857
https://news.ycombinator.com/item?id=38706074
not really. AI porn will never take off because ppl want to see a real person.
I've seen a lot of "this is impressive" but I'm not really seeing it. This looks to suffer from all the same continuity problems other AI videos suffer from.
What am I looking at that's super technically impressive here? The clips look nice, but from one cut to the next there's a lot of obvious differences (usually in the background, sometimes in the foreground).
As a gauge for how seriously I should take your critique:
How many hours a week are you actively using AI tools yourself?
What percentage of public comments that you’ve made about AI tools have been skeptical or critical?
> How many hours a week are you actively using AI tools yourself?
2 or 3. Mostly LLMs to check code.
> What percentage of public comments that you’ve made about AI tools have been skeptical or critical?
Probably around 90%.
So sell me. Why is this super impressive? I'm happy to admit that I'm pretty pessimistic about AI.
I have an eye for continuity issues, they are pretty obvious to me. Am I just too focused on that sort of a thing?
[flagged]
>It’s fucking video made by a computer after you type a sentence
Geez. Breathe a little bit. It's always weird to see somebody who had zero involvement in the creation/engineering/design of a product so disproportionately defensive about it.
The concept of text to video since LTX, CogVideo, AnimateDiff is closing in on two years of development at this point so there's naturally going to be a little less breathless enthusiasm.
If you had experience with even the locally hostable stuff like Hunyuan, Wan 2.2, VACE, etc. you'd probably be less impressed as well. The video they demoed had more fast cuts than a Michael Bay movie - illustrating the exact problem that video models STILL suffer from - a failure to generate anything longer than 60 seconds. In fact I didn't even see anything longer than 10 seconds at most. Maybe it's tailormade for an ADHD audience who grew up on Vine Videos.
On a more positive note, the physics have definitely improved though you can tell in some of the shots that the coherency degrades the longer scene goes on (see the volleyball clip).
> It's always weird to see somebody who had zero involvement in the creation/engineering/design of a product so disproportionately defensive about it.
Why is this weirder than the guy who goes out of his way to post negatively about them 90% of the time?
I'm assuming you've played more with AI video creation than I have.
Was there anything impressive here or is this mostly "meh"? I didn't see this solve any of the problems with AI videos but maybe this is solving something that I didn't know was a problem?
That's really what I'm trying to figure out with this announcement. Seeing 100s of comments about how impressive this is with no comments really discussing why has me trying to figure out what part of the hype I'm missing.
Good question. The one thing that Altman really seemed keen to play up was the whole integrate yourself into the video which from what I watched is definitely a step beyond the more conventional Image-To-Video models.
Depressingly that's probably a killer feature since if there's one thing people want to see more of it's themselves.
IMHO I also think the fact that they're trying to position themselves as a sort of infinite doom-scrolling tiktok lends support to the idea that their models are still only suitable for relatively short videos since coherency probably falls off a cliff after 30-60 seconds.
I think a lot of it is 'a dog trained to play the piano at a 4th grade level' effect: yes, it's really notable and impressive, but it gets old fast and isn't obviously useful beyond the novelty. You need to put in a lot more effort to get any actual value out of these tools and at that point their strengths and limitations become a lot more clear and it's obvious that they are not an everything machine.
I agree. But also I'd add that I'd be impressed and wowed if what you showed is "Actually, now the dog is playing piano at 5th grade level" or "Now the same dog can play the drums!".
Getting a "WTF are you talking about, it's a dog playing the piano, why aren't you impressed!" doesn't help the case. I've already seen a dog play the piano and this doesn't appear to be any better than the last time I saw it.
I'm not seeing comparisons of this to other video generation that already exists. That's what I want to see. Is this faster, more crisp, better at continuity? Why is this impressive vs competition?
> it's obvious that they are not an everything machine
Whoever said they were an everything machine?
When I first used the Internet in the 90s, I wasn’t like, “This sucks there’s no Amazon and no Facebook.”
I spend about 16 hours a week making AI videos. I'm not at all impressed by Sora 2. Open AI merely seems less embarrassing behind than they were, but still clearly behind.
As someone who loves seeing videos that appear after I type something, is there anything I should be noticing here? Or is this just bit of a special moment for people who feel, for some reason, there is some value for themselves personally in being deeply bought into OpenAI ecosystem and who aren't aware of the AI video scene as a whole?
I agree. It's really interesting that computers can make videos from sentences.
This, however, isn't the first AI capable of doing that correct? Didn't Sora 1, Veo, and others come out before this making videos from sentences?
Surely what makes this impressive isn't "This does what other people have done, including us Open AI".
Since you edited, I will to respond to your inflammatory edit
> Shocker lol
> Like a guy who hates tomatoes, bread, cheese, and pepperoni going, “This pizza sucks.”
This is a technology, don't reduce it down to preferences. There are obvious flaws with the video generating tech and one of the most annoying parts of talking to AI enthusiasts is the fact that they are unable to engage in honest dialog. All AI is amazing alien tech, it's always flawless and perfect.
What I've seen with AI video in the past is that they can make impressive on first glance looking videos but when you dig into them things are "off". Further, the continuity only lasts for 1 continuous shot. That creates videos where every 5 seconds you see a new shot and in the same setting things tend to drastically change.
Sora 2 appears to have all those problems, it doesn't appear to have solved any of them. That's why I ask "What's super impressive about this".
The same way I'd ask "What's super impressive about ChatGPT 5 vs 4". Snarkly saying "What are you talking about, it is a fucking realistic chat that can write a short story!" doesn't convince or impress me or anyone else that's a skeptic.
I'm not enthralled by AI. I'll happily use it when it makes sense and as it improves I'll probably use it more. For this, I don't see a significant improvement over the prior state of art.
> This, however, isn't the first AI capable of doing that correct? Didn't Sora 1, Veo, and others come out before this making videos from sentences?
It seems like there was a 90% chance you weren’t impressed by these either…
> doesn't convince or impress me or anyone else that's a skeptic
Yes, this is why I asked what % of comments you’ve made have been skeptical about AI. There’s nothing that will impress you.
I didn’t want to convince you I wanted to find out if I should have any interest in what you had to say.
> It seems like there was a 90% chance you weren’t impressed by these either…
And you'd be wrong in that assumption. When these first came out, particularly veo, I was quiet impressed with the photo realistic scenes it produced. Just like I was pretty impressed by "this is not a real human" website when it first launched.
> Yes, this is why I asked what % of comments you’ve made have been skeptical about AI. There’s nothing that will impress you.
New and novel things impress me. What's new and novel about this?
> I didn’t want to convince you I wanted to find out if I should have any interest in what you had to say.
Ah, so my opinion about stuff is only worth something if I said "I believe AI is the best thing that has ever been invented and I will kill myself to further it!!"
Go find a church if you can only talk with zealots.
> New and novel things impress me
They don’t seem to if 90% of what you have to say about AI is by your own admission at best tepid.
I honestly don’t understand how anyone can look at the state of AI today (even if the gains are perceived to be marginal compared to the last version) and be like, “meh, not impressive, better say something negative about it.”
But that seems to be the gospel here on the Church of HN.
The personas seem to be something new and pretty impressive.
You obviously never actually used Sora.
These are hand selected from thousands of prompts. The person who doesn't use these tools thinks that you prompt the video like you are Scorsese directing a film. That is exactly what AI video is not.
More like you write a prompt and then try try try to get something that vaguely does what you wanted it to do. Most of the time though it doesn't even get close. The best video is usually stuff you didn't even intend. It is great for demo reals but that is about it.
They already made such a big deal about Sora 1 for months before it came out then what have you seen from it after it actually released? Nothing. Absolutely nothing.
Honestly, Midjourney blows away what I seen in this video as far as being pretty but Midjourney video is the same problem. Your imagination is filling in all these AI video features that don't exist.
I have used Sora many times! If you need to see a Scorcese movie as output in order to be impressed then you will be incredibly disappointed.
> It is great for demo reals but that is about it.
This is what is amazing! Imagine you go back to 2020 and say, “Hey I’ve got this website where you can type and it’ll return back a little computer generated video that looks pretty real.”
The only thing lamer than people talking about how unimpressive AI is are the people talking about AI’s environmental impact.
Even someone who don't like pizza will recognize pizza is food.
Did that gauge even make sense?
If you're already a heavy user of AI tools, you've seen or used previous generations already. So it's just a gradual improvement, nothing to get excited about.
Just like smartphones have been incredibly boring in the last 10 years because the only change has been "slightly more performance" or "marginally thinner".
Really impressive engineering work. The videos have gotten good enough that they can grab your attention and trigger a strong uncanny valley feeling.
I think OpenAI is actually doing a great job at easing people into these new technologies. It's not such a huge leap in capabilities that it's shocking, and it helps people acclimate for what's coming. This version is still limited but you can tell that in another generation or two it's going to break through some major capabilities threshold.
To give a comparison: in the LLM model space, the big capabilities threshold event for me came with the release of Gemini 2.5 Pro. The models before that were good in various ways, but that was the first model that felt truly magical.
From a creative perspective, it would be ideal if you could first generate a fixed set of assets, locations, and objects, which are then combined and used to bring multiple scenes to life while providing stronger continuity guarantees.
"open ai is so nice because they spoon feed us little pieces of dog shit every few days to acclimate us to swallowing huge quantities of dog shit every single hours of your life in the near future, praise our benevolent god Sam Altman", and you should cheer for it!
> I think OpenAI is actually doing a great job at easing people into these new technologies. It's not such a huge leap in capabilities that it's shocking, and it helps people acclimate for what's coming. This version is still limited but you can tell that in another generation or two it's going to break through some major capabilities threshold.
This is a truly _wild_ way to describe "this version isn't much better than the previous one". Would you say "Apple's latest iPhone is a pretty small marginal improvement over the previous one, but it's useful to help peopel to acclimate for what's coming".
Impressively high level of continuity. The only errors I could really call out are:
1/ 0m23s: The moon polo players begin with the red coat rider putting on a pair of gloves, but they are not wearing gloves in the left-vs-right charge-down.
2/ 1m05s: The dragon flies up the coast with the cliffs on one side, but then the close-up has the direction of flight reversed. Also, the person speaking seemingly has their back to the direction of flight. (And a stripy instead of plain shirt and a harness that wasn’t visible before.)
3/ 1m45s: The ducks aren't taking the right hand corner into the straightaway. They are heading into the wall.
I do wonder what the workflow will be for fixing any more challenging continuity errors.
The video was slam cut together to avoid continuity problems. There was a lot of fast camera motion and unconnected scenes.
Particularly bad was the snowmobile sequence. It was literally a different snowmobile in every cut.
The racing pool duck scene was a different pool in every shot.
About the only consistent thing was the faces that were spliced into the scenes.
I do not really see anything super significant in the demo. It looks like this suffers from all the same problems of AI generated video. They just hid it by avoiding more then 5 seconds in the same setting.
The whole pool the ducks are racing at is a completely different pool when Sam starts talking.
The snowmobiles were different in each cut. The shape, color, and style of the lights were different.
The fact that this is their demo to the world and it's full of errors implies that average users will only get worse results.
I’m wary of being that damning, this early. What I want to know is, should my video have these kinds of continuity errors, how easily can I fix them?
It’s ok for this to be a fun toy. (And fun toy while also being an astonishing piece of engineering.) But if it wants to push beyond fun toy then it would be interesting to see how that process works.
Will Sora2 help me sketch out a movie for me, doing 10% of the work where I have to reshoot the other 90% for real, or will it get me 90% there leaving me only 10% left to do “by hand”?
(This is the exact same question, I believe, which is being asked of the maintenance burden imposed by vibe coded products. They get you 90% then fail spectacularly leaving you having to do the bulk of the work again? Or they get you 90% of the way and you int have to fill in the gaps to reach a stable long term product?)
I don't see how this is usable for making like a feature film. Editing will be impossible. At best it will be for ads. At worst for making social media slop.
It is not even just the errors. These video models are really impressive as long as you don't actually have something in your head you want to make. Then the laughable limitations are on full display.
I will believe it when I see because Sora 1 is probably the most disappointing technology given what I thought it was going to be that I can even think of. I waited forever for it and then barely used it because it sucks.
Not sure if it counts as a continuity error, but in the example "Prompt: Martial artist doing a bo-staff kata waist-deep in a koi pond", his wooden staff changes shape several times, resembling a bow at points. That was the first example I noticed as "clearly AI."
The Bo staff in the koi pond also seems to involve some impossible wrist movements
It definitely seems like a state-of-the-art model, but the sound having an underwater effect is the biggest tell. How long can it keep a scene going without things falling apart?
Very first frame of the video: green digital text is messed up. Stopped watching after that :)
I wonder about the implications of this tech.
State of the things with doom scrolling was already bad, add to it layoffs and replacing people with AI (just admit it, interns are struggling competing with Claude Code, Cursor and Codex)
What's coming next? Bunch of people, with lots of free time watching non-sense AI generated content?
I am genuinely curious, because I was and still excited about AI, until I saw how doom scrolling is getting worse
> just admit it, interns are struggling competing with Claude Code, Cursor and Codex
They are not. This is false, zirp ended, this is the problem. Not LLMs.
Of course primary cause could be ZIRP, but AI definitely accelerated the problem.
Interns at big tech maybe impacted less, because their systems are so complex, but when I look at job boards or talk with engineers I see they're mentioning interns less, AI assisted coding more.
Bar for the interns is higher now, why do I need 3 interns to polish the product if I can complete 70% of the job with AI and hire 1 intern to fix other parts
I know from a dev bootcamp that you are certainly wrong.
However, I also think ai coding is hyped way beyond its capability.
> dev bootcamp
i will not comment any further
Not sure your reply warrants any further expenditure of effort on my part, but for the benefit of other readers:
The bootcamp (actually, evening classes in coding run in cooperation with the public sector) regularly placed graduates with employers.
They’ve seen a big hit in this since AI, and companies have explicitly cited the fact that AI can complete the same tasks that these junior devs used to perform.
I’m wondering how they really prevent uploads of other peoples faces if they take a clip of a video of another person. I’m sure Apple didn’t open up the 3d Face ID scanning to them to verify
Try to provide some evidence first that AI is replacing people and that interns are struggling to compete with an LLM.
>What's coming next? Bunch of people, with lots of free time watching non-sense AI generated content?
Wasn't this always the outcome of the post labor economy?
For this discussion lets just say that AI+Robots could replace most human labor and thinking. What do people do? Entertainment is going to be the number one time consumer.
Paid by what?
Next comes SoraT.
Impressive tech. Don't love the likely societal implications.
It's obvious there is no way OpenAI can keep videos generated by this within their ecosystem. Everything will be fake, nothing real. We are going to have to change the way we interact with video. While it's obviously possible to fake videos today, it takes work by the creator and takes skill. Now it will take no skill so the obvious consequence of this is we can't believe anything we see.
The worst part is we are already seeing bad actors saying 'I didn't say that' or 'I didn't do that, it was a deep fake'. Now you will be able to say anything in real life and use AI for plausible deniability.
I think that's the point... Then world coin comes to the rescue
World coin is so delightfully dystopian. You could drop it wholesale into a superhero movie and it would be believable as the supervillain’s plot.
It's not that obvious. iOS is pretty secure, if they keep the social network and cameo feature limited to that there might not be good ways to export videos off the platform onto others beyond pointing a camera at the tablet screen. And beyond there being lots of ways to watermark stuff to be detectable, nothing stops the device using its own camera to try and spot if it's being recorded. The bar can be raised quite high as long as you're willing to exclude any device that isn't an iPhone/iPad.
Record things with 2 cameras.
Today's Sora can produce something that resembles reality from a distance, but if you look closely, especially if there's another perspective or the scene is atypical, the flaws are obvious.
Perhaps tomorrow's Sora will overcome the the "final 10%" and maintain undetectable consistency of objects in 2 perspectives. But that would require a spatial awareness and consistency that models still have a lot of trouble with.
It's also possible we remain stuck in the uncanny valley forever, or at least for the rest of our lives.
It's possible to produce some video or image that looks real, cherry-picked for a demo, but not possible to produce any arbitrary one you want that will end up passable.
>Everything will be fake, nothing real. We are going to have to change the way we interact with video.
I'm optimistic here.
Look at 1900s tech like social security number/card, and paper birth certificates. Our world is changing and new systems of verification will be needed.
I see this as either terribly dystopian - or - a possibility for the mass expansion of cryptography and encrypted/signed communication. Ideally in privacy preserving ways because nothing else will make as much sense when it comes to the verification that countries will need to give each other even if they want backdoor registry BS for the common man.
Breaking changes get fixes.
We are going to have to change the way we interact with video.
I doubt it will be for the better. The ubiquity of AI deepfakes just reenforces entrenchment around "If the message reinforces my preconceived notion, I believe it and think anyone who calls it fake is stupid/my enemy/pushing an agenda. If the message contradicts my preconceived notion, it's obviously fake and anyone who believes it is stupid/my enemy/pushing an agenda.". People don't even take the time to think "is this even plausible", much less do the intellectual work to verify.
They can't even be consistent within their own launch video. Consistency is by far the biggest issue with generative AI. How can a professional studio work with scenes which has continuity errors on every single shot? And if it's not targeting professionals, who is it for?
The common thread I am seeing with replacing creative work with AI is that of lowering the bar of acceptability and counterbalancing that with the (potential) savings from taking human labour out of the equation. The cost of labour is not just the raw cost but also the bargaining power that they can exercise by going on strikes etc. From my limited understanding, creatives seem to have more unions than programmers given that I have heard of at least two strikes from voice actors and writers and none from the tech sector. So it should be a win-lose for those who profit off of videos without taking part in the labour process of making one and lose-lose for everyone else.
> is that of lowering the bar of acceptability
Yes.
> counterbalancing that with the (potential) savings
No. It's all about personalization. Even with all the money in the world you couldn't sit a filming crew, VFX specialist, foley artist, and voice actors next to every user of your app, ready to produce new content in 60 seconds.
I don't get why this keeps being framed as a labor thing, it's unlocking genuinely new forms of interactive media.
What kind of personalisations are you hoping to see with this tech?
> I don't get why this keeps being framed as a labor thing
It's inextricably linked with labour. That doesn't mean that labour is only factor but it's an important one nonetheless.
You write a sentence and get out media... what more personalization are you looking for?
And no, labor is not a factor in the way you tried to frame it.
There is absolutely no one tying up $250,000 in GPUs to let users spit out a funny clip of Sam Altman jumping over a chair because they think that's a smart way a way to get out of paying artists.
>I don't get why this keeps being framed as a labor thing, it's unlocking genuinely new forms of interactive media.
Because it directly impacts people's ability to earn a living. If you truly don't understand this, I think you should spend some time talking to people who are impacted by it. Artists, and so on. Seriously, this is a head-in-the-sand take.
It's very bizarre to act this is being done in order to replace artists.
I build gen AI for entertainment: I don't build to replace anyone, and if my product gets eyeballs existing creators can't, it's because it gives the consumer something they wanted to see in the world.
Past that you're just complaining that consumers don't want what you made.
Myopic view. Just because your motives may be pure and your greed may be a non-issue doesn't mean that multinational corporations who have aligned themselves with fascist regimes (Meta, Alibiba, etc -- point is that they're demonstrably amoral/immoral) do not intend to exploit the tech to its fullest extent.
However having said that, the intention/aim need not be to deliberately replace creatives. That is not the claim I am making or that anyone in this thread or general public discourse is claiming. The minimal claim is simply that the commodification of art decreases the value and employability of people who perform the same task as the AI. It is also not limited to artists. It is all-encompassing. If an employer can now use AI instead of a copywriter, they will often do that -- big and small business alike. The same can be said for many niche fields which previously required specialized education or training.
I am not saying this from an Anti-AI perspective. I own an AI startup.
But use your big boy brain. It is clear that the commodification of intelligence has a downward market pressure on the market value (which is the PERCEIVED value of an employee in the eyes of the EMPLOYER) for many, if not eventually ALL jobs/roles.
I like how you start with "myopic view" the tell me to use my "big boy brain" to see the ease of creation across the board going down as a negative thing.
The purpose of the society was never to pay money to people, and if we figure out how to get grains of sand to replace skilled labor, there's no amount of greed that can outpace what it will do for humanity.
You are making a straw man argument. I am all for the benefits of AI; I can be pro-AI while still acknowledging the negative impacts.
OpenAI apparently assumes that the primary users of Sora 2/the Sora app will be Gen Z, especially with the demo examples shown in the livestream. If they are trying to pull users from TikTok with this, it won't work: there's some nuance to Gen Z interests than being quirky and random, and if they did indeed pull users from TikTok then ByteDance could easily include their own image/video generators.
Sora 2 itself as a video model doesn't seem better than Veo 3/Kling 2.5/Wan 2.2, and the primary touted feature of having a consistent character can be sufficiently emulated in those models with an input image.
Good point. I think OpenAI lacks the cultural understanding that tiktok is providing their users not only with entertainment but also social things like trends, reviews, gossip, self-expression. These aspects are not included in the sora experience.
This is going to sound crass but idc - OAI is just full of geeks, when what is needed is people who are more akin to hippies - thats pretty much what Apple was in the early days.
Its no use building technology when its not married with the humanities and liberal arts.
IMO you're making a valid point, because there seems to be a disconnect between AI and tangible human benefits. The ChatGPT-as-therapy train has been nerfed after the bad publicity, and it is force-fed to people at their workplaces through copilot.
I assume if you ask normal people how AI affects their lifes they'd think about annoying callcenter menus, deep fake porn and propaganda videos, and getting homework done. Not sure if any of this is a positive experience for the mind.
It's 2025 and most speech controls for car navigation don't work, Siri is a pile of sh*t and millionaires are trying to convince us that we should either use their AI or a google which has significantly reduced the quality of their search result pages.
It's like a false choice dilemma which allows back-to-the-roots companies such as Kagi to emerge, and I'm happy about it.
My comment, to my surprise, has received a lot of up-votes lol.
Completely agree. The way I think about life is - how will people look back 50 years from now, and make remarks about what is happening?
Physics seems better than veo 3 at least from demo videos
Sheeeeeeeeeeesh. That was so impressive. I had to go back to the start and confirm it said "Everything you're about to see is Sora 2" when I saw Sam do that intro. I thought there was a prologue that was native film before getting to the generated content.
I'm sorry but that's a gross exageration. If any of this was real film then I'd start a gofundme page for OpenAI to get better video production equipment and team because that would be laughably bad.
If anything, it looks a lot worse than a lot of AI-generated videos I've seen in the past, despite being a tech demo with carefully curated shots. Veo 3 just blows this out of the water for example.
It's not an exaggeration to me? I literally stopped the video and went back to the start and re-read. You're more than welcome to speak about your opinions and experiences, but I'm speaking about mine.
I'm over here thinking, "It felt like just yesterday I was laughing at trippy, incoherent videos of Will Smith eating spaghetti."
I love the progress we're making. I love the competition between big companies trying to make the most appealing product demos. I love not knowing what the tech world is going to look like in six months. I love not thinking, "Man. The Internet was a cool invention to have grown up in, but now all tech is mundane and extractive." Every time I see AI progress I'm filled with childlike wonder that I thought was gone for good.
I don't know if this represent SOTA for video generation. I don't care. In that moment I found it impressive and was commenting specifically on the joy I experienced watching the video. I find it frustrating to have that joy met with such negativity.
Don't worry. AI is going to be monetized and extractive in no time. Just like Social Media went from "fresh, fun and cool new tech" to "how did we let this horrible beast take hold of the world," AI will take the same path. In 10 years or sooner, when 99.99% of what you read, hear, and watch is AI slop, you're going to post "This used to be a cool invention!" if there's even a place left for humans to post by that time.
I agree. It will absolutely get there. Such is the trend of all scientific inventions. A breakthroughs occurs, prosperity follows in response, hedonic adaption causes satisfaction to regress to the mean, and then people squeeze every remaining drop of value out of the technology while we wait for those capable of true innovation to work their magic once more. I don't find it idyllic, but I accept it as the way the world works. It feels like a force of nature to me.
The period we're in is fleeting. I think it should be acknowledged and treasured for what it is rather than viewed with disdain because of what is inevitably to come. I stopped using Facebook and never moved to Insta/TikTok when things began to feel too extractive, but, for a good decade there, I felt so close to so many more people than I ever thought possible. It was a really nice experience that I no longer get to have. I'm not mad at social media. I'm happy I got to experience that window of time.
Right now I'm very happy to be using LLMs without feeling like I'm being preyed upon. I love that programming feels fresh and new to me after 15 years. I'm looking forward to having my ability to self-express magnified ten-fold by leveraging generative audio/visuals, and I look forward to future breakthroughs that occur once all these inventions become glorified ad-delivery mechanisms.
None of this seems bad to me. Innovation and technological progress is responsible for every creature comfort I have experienced in my entire life. People deserve to make livings off of those things even if they weren't solely responsible for the innovation.
I fully understand the hype but the initial scene with Sam feels nothing like how any self respecting video producer would create. The jump cuts mid-sentence are extremely jarring, certainly not framed in any traditional sense, and he's almost entirely out of focus.
Points though for the completely expressionless line delivery, it completely nailed that.
[dead]
You obviously have never actually tried to make anything in AI video. It is a parlor trick.Maybe this is a big advance but the current state of AI video is a joke. It is only impressive if you don't actually make anything. It is impressive in a marketing release that is quickly forgot about.
Will Smith eating spaghetti is the dumbest most uncreative thing. You are impressed by it because it is a meme. It is stupid.
I didn't claim to have made anything with AI video. I'm just commenting on how rapidly things appear to be improving from an external viewpoint. We used to give AI crap for failing to generate an appropriate number of fingers on still imagery. Now we're watching multiple minutes of video to find handfuls of discontinuities. The goalposts have shifted pretty far in an exceptionally short amount of time.
I have no idea why you're so intent on coming across bitter about a fledgling technology. A few years ago this demo video would've been indistinguishable from magic. It will continue to improve.
I get what you're saying. I see it too. That said, a lot of people won't notice the flaws, especially with these fast, choppy cuts. By the time you realize the neck is way too long or whatever, its 2 cuts later.
It definitely doesnt look worse tbh. Its impressive stuff you cant get around that
Anecdotally, I forgot multiple times that I was watching AI generated content, and my partner tuning in and out of it asked me a few times if we were watching the demo (as opposed to the real video). We are both pretty sensitive to slop too. I think something has flipped here.
[flagged]
Let me tell you, I wish I was paid by OpenAI, but unfortunately I get paid a very average wage in a country where OpenAI do not operate.
You can look at my account history and see that there is absolutely no indication that I am paid for, or a bot, I just happen to be relatively enthusiastic and bullish on AI capabilities. My account is 5 years old.
I do not think this product from OpenAI will be good for the world, I think it is mostly a very bad thing. I just think it is impressive.
I encourage you to try make the most generous interpretation of my comment and try to consider that other people can have sometimes wildly different experiences to you, however hard that is to believe. The alternative is to accuse people of being paid for commentors, which benefits no one and just isolates you further.
To add more nuance to my original comment, yes if you look closely at the videos it is clear that they are AI generated, but as I was watching I 'lapsed' out of attention a few times and had to remind myself what I was looking at was generated. If I stopped and rewound a few times, yes of course I could notice the slop. But my point is that it's the first time I have lost the "I am watching slop" thought while watching AI video.
Hope that clears things up for you.
Edit: I just looked through your comment history, you are an extreme AI perma bear, and that's ok, that doesn't mean you're a paid for anti-ai lobbyist. Also every single comment you have made is extremely negative and inflammatory.
I encourage you to revise your communication methods and try to practice some kindness.
No, the average person just doesn't have a creative bone in their body. They will be impressed when they get Super Hero movie #1899 but this time by AI , woo hoo!
Now that will be progress!
I am creative and have released music with a large listenership. The other person in my original comment is a designer by trade. Anything else dismissive you’d like to say or do you enjoy leaving snarky comments?
Anyone with access able to confirm if you can start this with a still image and a prompt?
The recent Google Veo 3 paper "Video models are zero-shot learners and reasoners" made a fascinating argument for video generation models as multi-purpose computer vision tools in the same way that LLMs are multi-purpose NLP tools. https://video-zero-shot.github.io/
It includes a bunch of interesting prompting examples in the appendix, it would be interesting to see how those work against Sora 2.
I wrote some notes on that paper here: https://simonwillison.net/2025/Sep/27/video-models-are-zero-...
Yes, you can start with a still and a prompt
One use that occurred to me is that fans will be able to "fix" some movies that dropped the ball.
For example, I saw a lot of people criticizing "Wish" (2023, Disney) for being a good movie in the first half, and totally dropping the ball in the last half. I haven't seen it yet, but I'm wondering if fans will be able to evolve the source material in the future to get the best possible version of it.
Maybe we will even get a good closure for Lost (2004)!
(I'm ignoring copyright aspects, of course, because those are too boring :D)
Just yesterday I learned "This summer, two Dramione fics turned rewritten novels became New York Times bestsellers" - https://slate.com/culture/2025/09/alchemised-senlinyu-harry-...
100% sure we will see people re-doing movie parts. Also see https://en.wikipedia.org/wiki/The_Phantom_Edit
> Maybe we will even get a good closure for Lost (2004)!
Whether it's text or super-advanced VR holograms, if it's fan fiction it's fan fiction. Which can be interesting and compelling, but that will never be as exciting as the Word of God[0]. Death of the Author is a nice thought experiment but few people really adhere to it, I've found.
0. https://tvtropes.org/pmwiki/pmwiki.php/Main/WordOfGod
Or just going to the Goofs section of a movie on IMDB, and fix the trivial issues (e.g. car had cracked window in earlier scene, and suddenly a normal window in another scene).
Much more mundane, but useful!
> (I'm ignoring copyright aspects, of course, because those are too boring :D)
You must understand that infinite copyright is the author's right, and AI companies must be sued for 50 trillion dollars.
Come on. This is just a fun thought exercise. I'm not suggesting creating a startup around this.
I was trying my hand at satire; but I understand that many people now genuinely hold such extreme views.
My issue is that the copyright aspect are what prevents me from using this as much as I otherwise would.
About 6 months ago I asked a few different AIs if they could translate a song for me as a learning experience, meaning not a simple translation, but more a word by word explanation of what each word meant, how it was conjugated, any more musical/lyrical only uses that aren't common outside of songs, and so on. I was consistently refused on copyright grounds, despite this seeming a fair use given the educational nature. If I pasted a line of the lyrics at a time, it would work initially, but eventually I would need to start a new chat because the AI determined I translated too much at once.
So in this one, if I wanted to ask it to create a video of the moment in Final Fantasy 6 when the bad guy wins, or a video of the main characters of Final Fantasy 7 and 8 having a sword duel, would it outright refuse for copyright reasons?
It sounds like it would block me, which makes me lose a bit of interest in the technology. I could try to get around it, but at what point might that lead to my account being flagged as a trouble maker trying to bypass 'safety' features. I'm hoping in a few years the copyright fights on AI dies down and we get more fair use allowance instead of the tighter limitations to try to prevent calls for tighter regulation.
Surely it wasn't deepseek right?
https://apps.apple.com/us/app/sora-by-openai/id6744034028
App link
edit: CBN80W for an invite code
I downloaded the app but I get a "Sora is invite only" screen after logging in to my OpenAI account and asking for an invite code.
> You can sign up in-app for a push notification when access opens for your account.
You need to be in the US/Canada and wait for this notification, and when you get an invite you can start using it in the app and on sora.com. And apparently you get 4 more invite codes that you can share with anyone, e.g. Android users:
> Android users will be able to access Sora 2 via http://sora.com once you have an invite code from someone who already has access
It's wild that I have a paid account but I have to scour the Internet to find someone else with a paid account and beg them for an invite code to use the product I already paid for. Make it make sense.
One thing that would make sense is for you to not pay any more.
But if you do, that signals to the company this is all perfectly okay.
This access code is "no longer available" :(
Check the browser console. The endpoint is returning 429 for me. So it might not even be accepting codes depending on how many you try.
Do you really want a "social" app for a firehose of high-fidelity slop?
If this is anything near the demo they have been released, this seems incredibly good at physics. Wow. Can't wait to try the new app.
Sora 1 was also lauded as being incredibly good at physics based on the early cherry-picked examples. The phrase "world simulator" was thrown around a lot. That didn't last long once people finally got their hands on it though.
Kind of wondersome if they will start to combine LLM generation with actual world models/GPU engines. Imagine that your model generates the wireframes, the Engine generates the physics and then another model fills in the actual visuals, and gaps... So you have realistic physics and gaps are filled in... Will also help with image retention more, if objects moved behind each other.
It was so much more hyped than that. They made it sound like Hollywood was in big trouble. It is going to have the same problems as Midjourney. You just don't have that much control of the scene. The process is to make thousands of random variations and cherry pick the good stuff because you can't do anything else.
The space dog and ice skater demo make it seem still very close to Sora 1
The demo on their homepage shows really bad physics. There’s a lot of it, but that doesn’t mean it’s correct. The hair of Sam looks like a paper cutout in almost every shot.
I received an invite and spent the last 6 hours generating an ungodly amount of free videos on the Sora app, they are AS GOOD as the cherry picked ones in the demo. They are insane. And the "social aspect" of the app is also very good.
I predict that this will move some people over, and IG/TT will lose marketshare.
Kling 2.5 is already pretty good at physics
I don't expect Sora2 to be SOTA. The Chinese models are further ahead in video/image gen
It's obvious there is no way OpenAI can keep videos generated by this within their ecosystem. Everything will be fake, nothing real. We are going to have to change the way we interact with video. While it's obviously possible to fake videos today, it takes work by the creator and takes skill. Now it will take no skill so the obvious consequence of this is we can't believe anything we see.
The worst part is we are already seeing bad actors saying 'I didn't say that' or 'I didn't do that, it was a deep fake'. Now you will be able to say anything in real life and use AI for plausible deniability.
I predict a re-resurgence in life performances. Live music and live theater. People are going to get tired of video content when everything is fake.
The Sora 2 livestream indicates that videos exported from the app will have visual watermarks.
Sure, then you just pump it through another model that removes watermarks.
In light of some comments and videos here, I’d like to morbidly announce that I can no longer distinguish between AI videos and real ones. However, I’ll take this as an opportunity to move from short-form content to long-form, since it seems that space hasn’t yet been hijacked by AI.
That could totally power next generation of green-screen techs. Generative actors may not find favorable response in the audiences; but SFX, decor, extras, environments that react to actors' actions - amazing potential.
There's less here than you think. Video games have already been procedurally generating environment art for quite some time, and film/tv are already leveraging that with giant screens that use Unreal Engine to create the backgrounds.
AI could be helpful here, but it's not clear that it is required or an improvement.
I can see that future generations are going to think that I'm boomer for preferring the performances of real actors instead of AI slop.
The music industry already went through this with AutoTune and we know how that turned out.
I don't get the autotune argument. It's like saying we shouldn't be using electronic instruments because it's not real or we shouldn't use digital audio instruments because they're not real etc.
It's just a way to get different kind of sound. It won't make you good tracks.
I think AI is starting to verge on making actual good music. The latest Suno release is wild.
An example here: https://v.redd.it/fqlqrgumo5rf1
I find this one interesting because Rap has classically been difficult for these models (I think because it's technically difficult to find the right rhythms and flow for a given set of lyrics).
I wonder what the prompt was for that.
The instrumental part is quite interesting but the lyrics/vocals...
it's just AI slop, like the median
like if you just put a bunch of words together and shipped that. Quantity was never what people wanted imo.
It is impressive if the instrumental track was made with just some prompts though
I actually think the vocals from ~2:00-~2:35 are pretty impressive there. It's wild to me that the models can play with tempo like that.
I've been listening to this across a variety of genres though, maybe these lyrics and vocals are more to your taste:
(similar to Opeth) https://suno.com/song/9ab8da05-c3f2-412d-80b4-c7d0b3ae840f?s...
(indie rock) https://suno.com/song/756dd139-4cba-4e40-b29c-03ace1c69673
I don't know but it doesn't impress me one bit? like I'm not trying to hate, but it just seems kind of like the model is given the track and then it tries to just follow it by matching words and then spitting them out, like as if it could talk about making a sandwich over some epic track and it'd sound the same?
like, LLMs are fantastic at generating patterns, so words that match and same with images etc.
But there's not much uniqueness? it's "impressive" like a savantic kind of ability to come up with rap, but it doesn't really product something I'd want to listen to..?
I listened to the metal thing and kind of the same thing?
It's very high fidelity, like the quality of the drums and etc it's quite impressive, but the vocals seem off? it's like a poem being read by TTS then transformed into "metal voice"
and kind of just an averaging of "metal music" kind of like stock photos and into a track, very formulaic
not to mention many metal bands etc they do formulaic stuff especially if they have an identifying kind of hit
But to me this is cool tech, but I wouldn't listen to it
I've listened music for a long time but I don't listen to a wide variety today, however for example with pop it can be very complex or very simple, but average or "almost" will really not make a good song, it can seem simple in hindsight but probably blood sweat and tears went into such songs, or creative energy that might never come back as strong.
just my raw thoughts though. it could be me being biased knowing it's AI, but I don't think so. I think my brain has kind of adapted to a point where I can feel if something is AI because it always seems super "average"/mid?
I'd love to see a blind study comparing a wide spectrum of these AI tracks to lesser known real artists (so the participants don't just recognize the songs) to see whether people can tell or if knowledge of the source biases them. I'm genuinely curious as to the results.
I don't think people would think anything strange of a lot of these tracks if they just randomly heard them on the radio.
When you listen to music that has been AutoTuned, you don't know if the singer can actually... sing. If you put them in a room and asked them to sing a song without artificial aid, would you actually enjoy their performance or not? You don't know!
This marked a divergence from thousands of years of vocal performances where singing ability and enjoyment of the music were one and the same.
AutoTune was the first slop, and the general population seems to like it.
The arguments against auto-tune are typically different, since it's obvious to anyone that autotune can't make you sound like a soprano if you're nowhere near - so skill is still required.
The problem with autotune is more that it removes a lot of nuance from singers' voices, it's like listening to MIDI instead of listening to a real piano. This is, however, something that can be improved. Synthesizers can produce wonderful musical effects, and there's lots of highly virtuoso music on synthesizers (including voice distortions, pretty similar technically to autotune) for those that are into it. Progrock, for example, was all about using new technology in complex and extremely interesting ways. Maybe more interestingly for your particular objection, you can look at early electronic music, say Vangelis or Isao Tomita or Kraftwerk. For at least parts of their songs, they could have just programmed their synthesizers ahead of time and played concerts without even being on stage - but that doesn't take away from the fact the music itself.
Ultimately, if the music sounds good and elicits some feelings and thoughts, it's good music. Whether the musicians can reproduce it live or it's done 90% in a studio doesn't really matter here. Of course, it does mean it may not be worth going to a live show from some particular performer, and it also means that the performer is not necessarily the most relevant artist - the person programming the "auto"tune should at least be considered part of the band.
That's like saying movies are not good cause they're not live action-only performances
For me the biggest thing is actually the production, there's many people involved usually and sometimes real magic gets made, and that magic might not even contain any vocals at first
like what is acceptable music? only raw vocals & acoustic instruments?
> The music industry already went through this with AutoTune and we know how that turned out.
Yeah, it turned out that almost all mainstream tracks nowadays have post-processing on vocals (the extent varying between genres and styles).
>The music industry already went through this with AutoTune and we know how that turned out.
they use it, everyone uses it, it got better to the point where most people dont know its used, ever heard of melodyne? well AI made it even better.
And then there has been about 20 years of people using it even as their style of music, notably in hip hop, reggaeton, urbano, country, etc.
Boomers like to think it was just an annoying fad in 2008-2011 or something, but it never went away, now everyone uses it, whether obvious or not
I don't understand why you think you'll be able to tell that far from now.
You can already do really cool stuff in this area “old” tech like stable diffusion. Not realistic or anything but really cool looking/morphing images
At least in terms of realism, the image generation field is at the realism line now. Single frame generation with Wan 2.1 / 2.2 (and others) for example, will get you realism.
Not related to Sora but, I have been looking for / hoping for an AI powered motion tracking solver. I've used Blender and Mocha in AE and both still require quite a bit of manual intervention, even in very simple scenes.
I saw some promnise with the Segment Anything model but I haven't seen anyone yet turn it into a motion solver. In fact I'm not sure if can do that at all. It may be that we need to use an AI algorithm to translate the video into a more simple rendition (colored dots representing the original motion) that can then be tracked more traditionally.
You should look at this Google paper (came out a few days ago):
https://video-zero-shot.github.io/
Since Agi is cancelled, at least we have shopping and endless video
Its almost like you need to have incremental steps that also generate revenue and push the technology forward
disparaging comments are also a motivator
Sam Altman has made (for me) encouraging statements in the past about short-form video like TikTok being the best current example of misaligned AI. While this release references policies to combat "Doomscrolling and RL-sloptimization", it's curious that OpenAI would devote resources to building a social app based on AI generated short form video, which seems to be a core problem in our world. IMO you can't tweak the TikTok/YouTube shorts format and make it a societal good all of a sudden, especially with exclusively AI content. This is a disturbing development for Altman's leadership, and sort of explains what happened in 2023 when they tried to remove him... -> says one thing, does the opposite.
>IMO you can't tweak the TikTok/YouTube shorts format and make it a societal good all of a sudden, especially with exclusively AI content.
I agree. At best, short videos can be entertainment that destroys your attention span. Anything more is impossible. Even if there were no bad actors producing the content, you can't condense valuable information into this format.
I'm optimistic about the Sora app! My hope is that it becomes much more whimsical and fun than TikTok because everyone on the app knows that all content is fake. Hopefully that means less rage-bait and more creative content, like OG YouTube. Nobody's going to get their news from Sora because it's literally 100% fake.
> it becomes much more whimsical and fun than TikTok because everyone on the app knows that all content is fake.
Sounds about as plausible as "ironically taking heroin".
> Nobody's going to get their news from Sora because it's literally 100% fake.
I'm with Neal Stephenson ("Fall", in this case) on this prediction, although I really hope I'm wrong.
That said… does anyone have an invite code?
Would also love an invite code if anyone has one.
> much more whimsical and fun than TikTok
In the early years everyone told me that TikTok is actually fun and whimsical (like just after it stopped being musical.ly), and it's all about fun collaboration, and amateur comedy sketches, fun dances and lipsyncs, and people posting fun reactions to each other etc, all lighthearted and that social media is finally fun again!
Why would it be more like OG YouTube, when the content they demoed very closely resembles YouTube shorts? The key difference is OG YouTube was long form.
> Hopefully that means less rage-bait
I have seen what people generate with AI, and I do not have good news for you.
Sam Altman is a businessman. His job is to say whatever assuages his market, and that includes gaslighting you when you're disgusted by AI.
If you never expected Altman to be the figurehead of principled philosophy, none of this should surprise you. Of course the startup alumni guy is going to project maligned expectations in the hopes of being a multi-trillion dollar company. The shareholders love that shit, Altman is applying the same lessons he learned at Worldcoin to a more successful business.
There was never any question why Altman was removed, in my mind. OpenAI outgrew it's need for grifters, but the grifter hadn't yet outgrown his need for OpenAI.
> His job is to say whatever assuages his market
I understand the cynicism but this is in fact not the job of a businessman. We shouldn't perpetuate the pathological meme that it is.
So the job of a businessman is not to increase shareholder value?
Nope. A CEO can't essentially steal from shareholders, but otherwise they have extremely broad latitude in how they engage in business.
There is no legal or moral imperative to make antisocial, unethical, or short term decisions that "maximize shareholder value."
This is something that morally weak people tell themselves (and others) to justify the depravity they're willing to sink to in order to satiate their greed.
The concept doesn't even make sense: different shareholders have different priorities and time horizons. A businessperson has no way to know what it objectively means to maximize their returns. They must make a subjective determination, and they have extremely broad latitude to do that.
If I run an AI business, then people using more AI means more business. If noone uses my AI then I go out of business
Increasing shareholder value can be done in the broadest sense by just increasing business
If I fund my own business, I can control growth and _choose_ ethics over profits, in the hope that stunting growth is acceptable if my customers value ethics too, and that whomever I someday pass my company to shares these values
If I take capital investment, I now have a contractual agreement to provide returns on that investment. Yes failure to adhere can result in lawsuits or legal penalties. Or I can be fired/voted out for failing to bring high enough returns. I now _cannot_ choose ethics over profits, due to the conflict of interest of self-preservation
So you are correct - there is no legal or moral contract to behave unethically, but there is instead a strong systemic and self-preserving incentive to do so
I think we almost agree here, but you make it sound as if the exec can simply stand up and do the right thing here. I argue the exec will simply be pushed aside for another
This is what people refer to when they talk about the binds that hold modern day mega-corps
If you yourself are an exec, I personally think you can understand these truths and work with them as best you can, and still be a good human being of course, but that there are lines that should not be crossed to keep a job
It is a collective issue we need to solve that of course starts with each individual seeing the true situation with kindness and compassion
You’re just saying there are incentives for unethical behavior? Yeah, obviously.
They don’t need to be excused by “well that’s their obligation.” It’s not! Actually, a person’s obligation is to act morally even when there are incentives otherwise, which is approximately all the time for nearly every person.
This is something children learn (lest they be excluded from their society) yet Very Smart People in the upper echelons of the business world conveniently forget.
> If I take capital investment, I now have a contractual agreement to provide returns on that investment. Yes failure to adhere can result in lawsuits or legal penalties.
This is not true. If you've signed a contract that says anything like this, consider getting a real lawyer.
To be clear I'm not disgusted by AI in general, I'm disgusted by short form video and AI/ML in service of dopamine reward loop hacking.
Someone who doesn't follow the moving edge would be forgiven for being confused by the dismissive criticism dominating this thread so far.
It's not that I disagree with the criticism; it's rather that when you live on the moving edge it's easy to lose track of the fact that things like this are miraculous and I know not a single person who thought we would get results "even" like this, this quickly.
This is a forum frequented by people making a living on the edge—get it. But still, remember to enjoy a little that you are living in a time of miracles. I hope we have leave to enjoy that.
Yeah. Just a few years ago, people here would have said stuff like that was decades away at best and pure science fiction at worst.
I remember having to tell people "Don't describe too many individual objects in your DALL-E 2 prompt, because the model has trouble separating discrete concepts and the features tend to blend together."
Now we have photorealistic video with sound and, oh yeah, the model can generate an entire script and mini-plot on its own based on the most basic prompt.
I know the comments here are gonna be negative but I just find this so sick and awesome. Feels like it's finally close to the potential we knew was possible a few years ago. Feels like a pixar moment when CG tech showed a new realm of what was possible with toy story
These videos are a very impressive engineering feat. There are a lot of uses for this capability that will be beneficial to society, and in the coming years people will come up with more good uses nobody today has thought of yet.
But clearly we also see some major downsides. We already have an epidemic of social media rotting people's minds, and everything about this capability is set to supercharge these trends. OpenAI addresses some of these concerns, but there's absolutely no reason to think that OpenAI will do anything other than what they perceive as whatever makes them the most money.
An analogy would be a company coming up with a way to synthesize and distribute infinite high-fructose corn syrup. There are positive aspects to cheaply making sweet tasting food, but we can also expect some very adverse effects on nutritional health. Sora looks like the equivalent for the mind.
There's an optimistic take on this fantastic new technology making the world a better place for all of us in the long run, after society and culture have adapted to it. It's going to be a bumpy ride before we get there.
I actually wonder if this will kill off the social apps and the bragging that happens. It will be flooded by people faking themselves doing the unimaginable.
This is also my thesis. The internet is going to be saturated with AI slop indiscernible from real content. Once it reaches a tipping point, there will no longer be much of a reason to consume the content at all. I think social networks that can authenticate video/photo/text content as human-created will be a major trend in a few years.
I have no clue if the reactions are real, but there are some videos online of people showing their grandparents gameplay from Grand Theft Auto games trying to convince them that it is real footage. The point of the videos is to laugh at their reactions where they question if it really happened, etc.
Maybe this will result in something similar, but it can affect more people who aren’t as wary.
Right now with kids, the current trend is to prank their parents using Gemini into thinking they let a homeless guy in their house
https://www.tiktok.com/discover/ai-homeless-people-in-my-hou...
Heh, fast forward a few years and nobody’s surprised anymore when someone falls for a video which is the result of two sentences long instruction.
Depending on which internet you do mean, cause meta & insta are NOT THE Internet.
there will be billions of people consuming the content
But then you’re creating an incentive for the AI slop to become so realistic it is indistinguishable from actual video.
Unless there some fundamental, technical way to distinguish the two, I wonder who would win?
I regularly get AI movie recaps on my shorts and I just eat it up.
The very fact that I (or billions of others) waste time on shorts is an issue. I don't even play games anymore, it's just shorts. That is a concerning rewiring of the brain :/
Guess what I`m trying to say is that, there is a market out there. It's not pretty, but there certainly is.
Will keep trying to not watch these damn shorts...
there would need to be cameras that can cryptographically sign videos with trusted vendor keys, or perhaps there is some other solution.
This is what https://c2pa.org/ is for. I think some camera vendors already have support.
Yes, I wonder if the content distribution networks that call themselves "social networks" can even survive something like this.
Of course, the ones focusing on the content can always editorialize the spam out. And in real social networks you ask your friends to stop making that much slop. But this can be finally the end of Facebook-like stuff.
> There are a lot of uses for this capability that will be beneficial to society
Are there? “A lot” of them? Please name a few that will be more beneficial than the very obvious detrimental uses like “making up life-destroying lies about your political opponents or groups of people you want to vilify” or “getting away with wrongdoing by convincing the judge a real video of yourself is a deepfake”.
That last one has already ben tried, by the way.
https://www.theguardian.com/technology/2023/apr/27/elon-musk...
It can generate funny videos of bald JD Vance and Harry Potter characters for TikTok. Which makes me wonder, what is the actual plan to make money off these models? Billions have been invested but the only thing they seem to be capable of is shitposting and manipulation. Where is the money going to come from?
> There are a lot of uses for this capability that will be beneficial to society
Please enlighten me. What are they? If my elderly grandma is on her deathbed and I have no way to get to see her before she passes, will she get more warmth and fond memories of me with a clip of my figure riding an AI generated dragon saying goodbye, or a handwritten letter?
What about a new electric guitar? Your grandma wouldn't want that on her deathbed so it's useless? Cmon man.
Still zero responses, eh? My example was charged but I clearly had a point: how does AI fill a void where meaning should be, over what has worked for centuries? How is it better than face to face, or a handwritten letter?
I don't think anyone is saying it is.
> There are a lot of uses for this capability that will be beneficial to society
My original question was asking for examples of this. Try to keep up, c'mon man
Suno allows me to rapidly flesh out demos and brainstorm. Played music my whole life manually. Easier for me to find what I'm looking for and while avoiding demo love.
I still feel this is limited by what it learned from. It looks cool but it also looks like something I'd dreamt or saw flicking through TV channels. Kind of like spam for the eyes.
It looks like it has been trained exclusively on car advertisement videos playing at airports.
No doubt they can create Hollywood quality clips if the tools are good enough to keep objects consistent, example, coming back to the same scene with same decor and also emotional consistency in actors
> keep objects consistent
I think this is not nearly as important as most people think it is.
In hollywood movies, everyone already knows about "continuity errors" - like when the water level of a glass goes up over time due to shots being spliced together. Sometimes shots with continuity errors are explicitly chosen by the editor because it had the most emotional resonance for the scene.
These types of things rarely affect our human subjective enjoyment of a video.
In terms of physics errors - current human CGI has physics errors. People just accept it and move on.
We know that superman can't lift an airplane because all of that weight on a single point of the fuselage doesn't hold, but like whatever.
Water level in a glass changing between shots is one thing, the protagonist’s face and clothes changing is another.
Location consistency is important. Even something as simple and subtle as breaking the 180-rule [1] feels super uncanny to most audiences. Let alone changing the set the actor occupies, their wardrobe, props, etc.
There are lots of tools being built to address this, but they're still immature.
https://x.com/get_artcraft/status/1972723816087392450 (This is something we built and are open sourcing - still has a ways to go.)
ComfyUI has a lot of tools for this, they're just hard to use for most people.
[1] https://en.wikipedia.org/wiki/180-degree_rule
Well put. Honestly the actor part is mostly solved by now, the tricky part is depicting any kind of believable, persistent space across different shots. Based off of amateur outputs from places like https://www.reddit.com/r/aivideo/, at least!
This release is clearly capable of generating mind-blowingly realistic short clips, but I don't see any evidence that longer, multi-shot videos can be automated yet. With a professional's time and existing editing techniques, however...
People got used to James Bond actors changing between movies, but from scene to scene in the same movie would be a bit confusing.
It all depends on quantity and "quality" of the continuity errors. There's even a job for it https://en.wikipedia.org/wiki/Script_supervisor
I wonder if this stuff is trained on enough Hallmark movies that even AI actors will buy a hot coffee at a cafe and then proceed to flail the empty cup around like the humans do. Really takes me out of the scene every time - they can't even put water in the cup!?
No way man, this is why i loved Mr Robot, they actually payed a real expert and worked story around realism and not just made up gobbleygook that shuts my brain off entirely to its nonsense
Cool demo! But let’s pour one out for all the weird, janky, hand crafted videos that made early internet so fun. Anyone else still crave that kind of content?
I already get enough AI spam and scam videos on social media. I don’t need them to be better quality
Potential for what exactly? More of 30-sec slop?
Pixar moment for me means a novel techonology evoking a profound emotional response for the first time. This was not it.
The ability for the masses to create any video just by typing, among the other features, is not novel technology? Or is it just the lack of emotional response?
Yes, profound emotional response. There were cg animations before Pixar.
> Feels like a pixar moment when CG tech showed a new realm of what was possible with toy story
@qoez
> The first entirely AI generated film (with Sora or other AI video tools) to win an Oscar will be less than 5 years away.
https://news.ycombinator.com/item?id=42368951
This prediction of mine was only 10 months ago.
Imagine when we and if we get to 5 years.
This Sora 2 generation of Cyberpunk 2077 gameplay managed to reproduce it extremely closely, which is baffling: https://x.com/elder_plinius/status/1973124528680345871
> How the FUCK does Sora 2 have such a perfect memory of this Cyberpunk side mission that it knows the map location, biome/terrain, vehicle design, voices, and even the name of the gang you're fighting for, all without being prompted for any of those specifics??
> Sora basically got two details wrong, which is that the Basilisk tank doesn't have wheels (it hovers) and Panam is inside the tank rather than on the turret. I suppose there's a fair amount of video tutorials for this mission scattered around the internet, but still––it's a SIDE mission!
Everyone already assumed that Sora was trained on YouTube, but "generate gameplay of Cyberpunk 2077 with the Basilisk Tank and Panam" would have generated incoherent slop in most other image/video models, not verbatim gameplay footage that is consistent.
For reference, this is what you get when you give the same prompt to Veo 3 Fast (trained by the company that owns YouTube): https://x.com/minimaxir/status/1973192357559542169
> Everyone already assumed that Sora was trained on YouTube
Doesn't this already answer your question...? "Let's Play" type videos and streams have been a thing for years now, even for more obscure games. It very well could've been trained on Cyberpunk videos of that mission.
It's hard for me to believe that the model coherently memorized both the video and audio of a relatively obscure Let's Play, and that a simple prompt was enough to surface it (the use of the term "Basilisk tank" would also likely not be in video metadata either). That is the reason the person who made that tweet, who has far more prompting experience than myself, was shocked.
It’s hard for you to believe, sure, and I recognize the context of who tweeted it.
I still maintain that’s the kernel it’s getting it from. It’s impressive, I’m just not really shocked by it as a concept.
That's really interesting. What if they RAG search related videos from the prompt, and condition on that to generate? That might explain fidelity like this
An interesting counterexample is "a screen recording of the boot screen and menus for a user playing Mario Kart 64 on the N64, they play a grand prix and start to race" where the UI flow matches the real Mario Kart 64, but the UI itself is wrong: https://x.com/fofrAI/status/1973151142097154426
I like the player being in "1th" while being behind everyone else. Still crazy though.
Their anime vid gen is really, really impressive. The results I've seen aren't /good/ from an industry-standard (nothing compared to the likes of the Demon Slayer movie I watched in theatres recently), but I legitimately couldn't tell that it was AI-generated. Massive step up from Sora 1 and other vid gen models.
Here's to hoping that the industry will adapt to have it aid animators for in-betweening and other things that supplement production. Anime studios are infamously terrible with overworking their employees, so I legitimately see benefits coming from this tool if devs can get it to function as proper frame interpolation (where animators do the keyframes themselves and the model in-betweens).
I'm a software engineer and hobbyist actor/director. My friends are in the film industry and are in IATSE and SAG-AFTRA. I've made photons-on-glass films for decades, and I frequently film stuff with my friends for festivals.
I love this AI video technology.
Here are some of the films my friends and I have been making with AI. These are not "prompted", but instead use a lot of hand animation, rotoscoping, and human voice acting in addition to AI assistance:
https://www.youtube.com/watch?v=H4NFXGMuwpY
https://www.youtube.com/watch?v=tAAiiKteM-U
https://www.youtube.com/watch?v=7x7IZkHiGD8
https://www.youtube.com/watch?v=Tii9uF0nAx4
Here are films from other industry folks. One of them writes for a TV show you probably watch:
https://www.youtube.com/watch?v=FAQWRBCt_5E
https://www.youtube.com/watch?v=t_SgA6ymPuc
https://www.youtube.com/watch?v=OCZC6XmEmK0
I see several incredibly good things happening with this tech:
- More people being able to visually articulate themselves, including "lay" people who typically do not use editing software.
- Creative talent at the bottom rungs being able to reach high with their ambition and pitch grand ideas. With enough effort, they don't even need studio capital anymore. (Think about the tens of thousands of students that go to film school that never get to direct their dream film. That was a lot of us!)
- Smaller studios can start to compete with big studios. A ten person studio in France can now make a well-crafted animation that has more heart and soul than recent by-the-formula Pixar films. It's going to start looking like indie games. Silksong and Undertale and Stardew Valley, but for movies, shows, and shorts. Makoto Shinkai did this once by himself with "Voices of a Distant Star", but it hasn't been oft repeated. Now that is becoming possible.
You can't just "prompt" this stuff. It takes work. (Each of the shorts above took days of effort - something you probably wouldn't know unless you're in the trenches trying to use the tech!)
For people that know how to do a little VFX and editing, and that know the basic rules of storytelling, these tools are remarkable assets that compliment an existing skill set. But every shot, every location, every scene is still work. And you have to weave that all into a compelling story with good hooks and visuals. It's multi-layered and complex. Not unlike code.
And another code analogy: think of these models like Claude Code for the creative. An exoskeleton, but not the core driving engineer or vision that draws it all together. You can't prompt a code base, and similarly, you can't prompt a movie. At least not anytime soon.
Taking the time and effort out of something is exactly what strips it of its beauty
Beauty is not just an “idea” that someone has and needs to get out onto a medium
It is a process and journey that a person undergoes to get said idea onto said medium
That journey often plays out very differently than the person expects. Things change, the art is different from the idea, and the person learns and grows
Our modern society is so obsessed with results, competition, and efficiency that we no longer see the truth: the journey is to be enjoyed, and from enjoying the journey, comes beauty
I encourage you to meditate on why our society is so sick and depressed right now, and extrapolate to how we got here, before assuming this will be a good thing for society
I saw a quote earlier this week that I'll copy here:
> I considered renting out sound stages, flying to exotic desert locations, getting a scuba team to shoot the underwater scenes in an aquarium, commissioning custom-made Teletubbies costumes, hiring SAG actors, building dozens of miniature sets, and spending my life savings on making this video. But using AI just seems slightly easier.
Making short films with AI is still incredibly effortful. If you're being careful and diligent, it takes days to "shoot" and edit the entire shot list for a 5-7 minute short.
Would you say that the creators of today's animated TV shows, in mechanizing production with Toon Boom Studio, have stripped the beauty away? I still found "Bojack Horseman" to be a salient dramedy.
Would you say that Pixar, in using motion capture and algorithms to simulate light, physics, and movement, is cutting away the journey?
This is a new adventure and new level of abstraction we're embarking upon.
I'm already thinking about the next way points: real time mocapped improv for D&D campaigns and live community theater fantasy and science fiction productions.
These are tools that bring us to new places, that enable us to tell new stories. Previously you'd have to win Disney budget approval to tell a story matching your vision - now you don't.
But I will still be entertained. Expedient AI expression can touch most people the same way a low effort meme or an off the cuff whitticism.
Art is not effort. Art is not labour. Beauty is not suffering. Art =/= craft. Art is communication.
If someone wants to suffer long the endurance journey to becaome skilled at a craft we can still respect/appreciate it the same way a sprinter spends 10 years training to run real fast, in the mean time most of us will use a vehicle to get somewhere faster.
What we're going to lose is a bunch of interesting behind the scene videos because no one is going to watch someone prompt for an hour wondering why can't I do that, but rather why didn't I do that.
Proliferating tools for creation is net good in the same sense that teaching masses to write is net good. It's strange people are opposing lowering the barrier to entry to visual communication. That's what art ultimately is, communication. Once difficult, soon ubiquituous.
So landscapes are not beautiful?
You think landscapes are created in an instant?
Creative people with ambition and limited resources make good things today without this technology. All this does is accelerate the rate at which low quality "content" is produced by people that have no interest in learning a craft, without attribution and without compensation for the people that have made the effort and whose works train these models.
Now, creative people with ambition and limited resources have a new, powerful tool.
This will also be used to create great content.
We mustn't teach peasants how to read or write or else one day we'll live in hellscape of infinte unread inboxes and eternal september... both of which sucks, but much less than a world where masses were illiterate. Ai art/slop is just that for visual communication. Now it's supremely shitty this power is currently being monetized / controlled by a few, the same way communication material like papyrus and paper was jealously guarded/exploited before printing press proliferated, then the mistake that was the internet giving desemination power to every pleb. But there's free models out there and maybe (hopefully) in the not too distant from now the barrier to entry is accessible commodity level hardware and artists just have to eat shit and realize they've contributing to the creative common/canon like those before, i.e. world where monks/literati simply copied and duplicated work in pre copyright era because knowledge/expression, unless jealously guarded, was a collective resource to be built upon.
Precisely.
I have a really big problem with letting low quality stuff infest into the species.
I wrote this a year or so ago: https://www.technicalchops.com/articles/ai-goes-to-hollywood...
"The studios and creators who thrive in this new landscape will be those who can effectively harness AI’s capabilities while maintaining the human creativity and vision that ultimately drives the art of cinema."
It is in many ways thrilling to see this come to life, and I couldn't agree with you more.
> "The studios and creators who thrive in this new landscape will be those who can effectively harness AI’s capabilities while maintaining the human creativity and vision that ultimately drives the art of cinema."
..Just somehow several years on, these optimistic statements still all end up being in the future tense, somehow for all the supposed greatness and benefits, we still dont see really valuable outputs. A lot of us do not want more of the "CONTENT" as envisioned by corporate ghouls who want their employees or artists to "thrive" (another word kidnapped by LinkedIn-Linguists). The point is not in the speed and easiness of generation of outputs, visual and sound effects etc. The point is the artists interpretation and their own vision, impressions etc. Not a statistical slop which "likely" fits my preferences (i.e. increases my dopamin levels).
Adding to the list: The Adventures of Reemo Green. Very funny, and the first time I’ve watched AI video and enjoyed it as more than a technical curiosity.
https://www.youtube.com/watch?v=5bYA2Rv2CQ8
sorry but it's funny that you mention "heart and soul" while sharing some of the most soulless videos i've ever seen.
I'll have you know that in this year's Atlanta 48 Hour Film project (something I've been doing since I was a teen), several teams used AI.
Rewind to just one year prior -- 2024.
AI video was brand-spanking new. We'd only just gotten over the "Will Smith" spaghetti video and the meme-y "Pepperoni Hug Spot" and "Harry Potter by Balenciaga" videos.
I was the only person to attempt to use AI in 2024's competition. It was a time when the tools and infrastructure for video barely existed.
On the debut night, I was resoundingly booed by the audience. It felt surreal. Working all weekend to have an audience of peers jeering at you in a dark theater. The judges gave me an award out of sympathy.
Back then, image-to-video models really were not a thing (Luma launched "Dream Machine v1" shortly after this). I was using Comfy, Blender, Mocap, a full Mocap suit (the itchy kind), and a lot of other hacks to build something with extremely crude tools.
We lost a day of filming and had to scramble to get something done in just 24 hours. No sleep, too much caffeine. Lots of sweat and toil.
The resulting film was a total mess, of course:
https://vimeo.com/955680517/05d9fb0c4f (It's seriously bad - I hate it. It might legitimately be the very first time AI was used in a 48 hour competition.)
That said, it felt very much like a real 48 Hour competition to me. Like a game jam. The crude ingredients, an idea, the clock. The hustle. The corners being cut. It was palpable.
I don't think you can say there isn't soul in this process. The process has so much soul.
Anyway, fast forward to this year. Three teams used AI, including my own. (I don't think I have a link to our film, sadly.)
We all got applause. The audience was full of industry folks, students, and hobbyists. They loved it. And they knew we used AI.
The industry is anxious but curious about the tech. But fundamentally, it's a new tool for the tool box. The real task is storytelling.
Well I was entertained.
What is up with a lot of voices are left ear only?
Carter needs a new laptop. His daily driver has been falling apart for ages but he refuses to give it up.
We all told him about the sound mix - he let a couple of videos slip with a bad "mono as single-channel stereo audio" renders. On his machine it sounded normal. He got flack for that, and he's been hearing this for months.
I'm going to show him this thread. I don't think he'll ever forget to check again.
Despite that, he's a really talented guy. Chalk this up as a bad production deploy. We didn't want to delete and re-upload since the videos had legs when we first released them. There's a checklist now.
Lol I wish YT had a warning for that.
In the meantime, good old
Settings -> Accessibility -> Audio -> Play Stereo as Mono
helps.
My left ear enjoyed these videos
The voice quality in the generated vids is surprisingly awful.
That's the first thing I noticed, too. The first words you hear in the trailer sounds like someone ran the voice through a comb filter. It's so bad it made my skin crawl immediately.
I predict a re-resurgence in life performances. Live music and live theater. People are going to get tired of video content when everything is fake.
One would think, but people are spending less on live events due to costs
likely because we haven't yet reached peak slop/exhaustion by slop. Soon enough...soon enough
Buying lots of calls on Live Nation.
Most of human crafted shorts / reels are already slop.
There's something about the faces that looks completely off to me. I think it's the way the mouth and whole face moves when they talk.
It's the inaccuracy of things like shadows, sub-surface scattering and specular highlights. I think the shadow inaccuracy is what the human visual system is most sensitive to.
These LLMs might make content that looks initially impressive but they are absolutely not performing physically based rendering or have any awareness of the lighting arrangement in these scenes. There are a lot of things they get right, but you only have to screw up one small element to throw the whole thing off.
I am willing to bet that Unreal Engine 5 will continue to produce more realistic human faces than OAI ever can with these types of models. You cannot beat the effects of actually running raytracing in a PBR pipeline.
Yeah, the faces aren't right, and impressive as it is I'm getting icky "uncanny valley" vibes from this.
CGI for fantasy stuff is unavoidable, but when it's stuff that could have been done by actors but is instead AI, then to me it just feels cheap and nasty - fake.
Going to be an amazing source of training data, wait till they get it to real time and people are leaving their video camera open for AR features. OpenAI is about to have a lot of current real world image data, never mind the sentiment analysis.
I don't think they were limited for video training data. Gathering real world data is pretty easy, gathering curated information is a little more difficult.
IDK if the site is being hugged to death but I can only load the first video. Even in just one viewing there were noticeable artifacts, so my impression is that Veo is still in the lead here.
Yeah I am curious what the actual resolution of these videos will be. The launch videos on this link will only play in like 360p for me.
Last famous words:
> A lot of problems with other apps stem from the monetization model incentivizing decisions that are at odds with user wellbeing. Transparently, our only current plan is to eventually give users the option to pay some amount to generate an extra video if there’s too much demand relative to available compute. As the app evolves, we will openly communicate any changes in our approach here, while continuing to keep user wellbeing as our main goal.
Sam will quickly learn that general users give -zero- thought to OpenAI well being. Nor be bothered that they should give it a thought.
> And we're introducing Cameo, giving you the power to step into any world or scene, and letting your friends cast you in theirs.
How much are they (and providers of similar tools) going to be able to keep anyone from putting anyone else in a video, shown doing and saying whatever the tool user wants?
Will some only protect politicians and celebrities? Will the less-famous/less-powerful of us be harassed, defamed, exploited, scammed, etc.?
it seems like this is basically youtube's ContentID, but for your face. as long as you upload your "cameo" aka facial scan to them, they can recognize and control the generation of videos with it. if you don't give them your face, then they can't/won't.
"Consent-based likeness. Our goal is to place you in control of your likeness end-to-end with Sora. We have guardrails intended to ensure that your audio and image likeness are used with your consent, via cameos. Only you decide who can use your cameo, and you can revoke access at any time. We also take measures to block depictions of public figures (except those using the cameos feature, of course). Videos that include your cameo—including drafts created by other users—are always visible to you. This lets you easily review and delete (and, if needed, report) any videos featuring your cameo. We also apply extra safety guardrails to any video with a cameo, and you can even set preferences for how your cameo behaves—for example, requesting that it always wears a fedora."
Brilliant.
Until you have 2 people that are near identical. They don’t even have to be twins, there are plenty of examples where people can’t even tell other people apart. How is an AI going to do it?
You don’t own your likeness. It’s not intellectual property. It’s a constantly changing representation of a biological being. It can’t even be absolutely defined— it’s always subject to the way in which it was captured. Does a person own their likeness for all time? Or only their current likeness? What about more abstract representations of their likeness?
The can of worms OpenAI is opening by going down this path is wild. We’re not current able to solve such a complex issue. We can’t even distinguish robots from humans on the internet.
I'm an identical twin so immediately I can see a pretty stupid obvious problem with this.
If this company's guardrails end up sufficiently working well in practice (note phrases like "intended", "take measures", and "preferences...requested", on things they can't do 100%)... there will be weak links elsewhere, letting similar computation be performed without sufficiently effective guardrails against abuse?
How do we prepare for this? Societal adjustment only (e.g., disbelieving defamatory video, accepting what pervs will do)? Establishing a common base of cultural expectations for conduct? Increasing deterrence for abusers?
Looks like it requires you to film yourself from specific angles and while repeating an autogenerated phrase. Like a pre-AI selfie taken with a handwritten placard with your username and the date or whatever.
Basically deepfakes for everyone.
Normalizing effective abolishment of consent for imagery, or just consent in general for just about anything, when it can portray anyone doing anything.
Ah the great trade off that comes with little to no regulation.
Honestly this is the safest possible outcome.
If Deepfakes remain the tools of nation state actors, laypeople will be easily fooled.
If Deepfakes are available on your iPhone and within TikTok, everyone will just ask "Is it Photoshop?" for every shred of doubt. (In fact, I already see people saying, "This looks like AI".)
This is good. Normalize the magic until it isn't magic anymore.
People will get it. They're smart. They just need exposure.
> People will get it. They're smart. They just need exposure.
I really doubt this.
If you are in the creative field, your work will just be reduced to "is this slop?" or "fixed it!" with a low effort AI generated work of your original work (fuck copyright right?).
I already see artists battling and fighting putting out their best non AI work only for their audience to question if it is real and they lose the impressiveness.
This just already undermines creators who don't use AI generated stuff.
But who cares about them right? "it is the future" and it is most definitely AGI for them.
But then again, the starving artist never really made any money and this ensures that the artform stays dead.
> People will get it. They're smart. They just need exposure.
It's either this, or the opposite (eg, misinformation needs to be censored). Seems like we as a society can't quite make up our mind on which approach to take.
Clicking a link on the OpenAI dashboard and beeing greeted with a full page of scandily clad women was certainly not what I expected to see when opening Sora..
You think too highly of us (humans)
Ok that's technically really impressive, and probably totally unusable in a real creativity context beyond stupid ads and politically-motivated deepfakes.
Amazing. iOS only, with region restrictions in 2025.
> Sora is not available in Puerto Rico yet
I love the casual reminds that we're second-class citizens each time a new technology gets released. Available in the US but always excluding Puerto Rico.
considering legal foolishness of EU, this is the right move.
I welcome a world where gullible people begin to doubt everything they see.
The result of that isn't rational skepticism, though. It's distrusting mainstream news and embracing whatever conspiracy theories your friends believe.
Wait, when did mainstream news become trustworthy? =)
Distrusting all mainstream news indiscriminately (or whenever you feel like it) is at least as bad as trusting it indiscriminately.
My friend, no man ever got rich betting against the infinitely deep well of human gullibility.
I feel so bad for the climate now.
Doing this as a social app somehow feels really gross, and I can't quite put to words why.
Like, it should be preferable to keep all the slop in the same trough. But it's like they can't come up with even one legitimate use case, and so the best product they can build around the technology is to try to create an addictive loop of consuming nothing but auto-generated "empty-calories" content.
I see it more as recognizing that it will take time to be good enough for other use cases so for this release they are targeting it as just something to have fun with. After seeing LLMs crammed into everything whether it makes sense or not, I can appreciate that.
Latest Invite Codes Community-shared Sora invite codes updated in real-time
https://sorainvitecode.org/
What makes TikTok fun is seeing actual people do crazy stuff. Sora 2 could synthesize someone hitting five full-court shots in a row, but it wouldn’t be inspiring or engaging. How will this be different than music-generating AI like Suno, which doesn't have widespread adoption despite incredible capabilities?
It's hard to believe, but some people enjoy. On the other hand, some popular content on TikTok is probably worse than AI generated content and that's another problem...
Considering that much of the TikTok content you mention is staged or heavily edited, this skips the make-believe.
Just seeing the examples that I assumed are cherry picked, it seems like they're still behind on Google when it comes to video generation, the physics and stylized versions of these shots seem not great. Veo3 was such a huge leap and is still ahead of many of the other large AI labs.
I can see it being interesting to create wacky fake videos of your friends for a week or two, but why would people still be using this next year?
I watch videos for two reasons. To see real things, or to consume interesting stories. These videos are not real, and the storytelling is still very limited.
A lot of realslop is fake too. As in staged but pretended as real for rage bait or annoyance bait. Or stupid shaggy dog story videos, where it seems like the thing will happen any moment now and then nothing happens.
One recent disillusionment for me was that lots of police body cam content is fake, as in basically amateur actors trying to enact a realistic police stop, they even put the usual bodycam numbers and letters and axos logo in the corner etc.
And so many other videos of things happening in the street are more or less obviously fake and staged. Still 90% probably don't notice.
I'm no Nostradamus, but I predict these models will be much better in a year.
In the right hands it's a new art medium. Some (few, maybe) midjourney generations are serious art.
So, for the same reason you'd go to a local art gallery
soft porn
You only watch real things? Have you never watched a movie?
> or to consume interesting stories
> Concerns about doomscrolling, addiction, isolation, and RL-sloptimized feeds are top of mind—here is what we are doing about it.
> We are giving users the tools and optionality to be in control of what they see on the feed. Using OpenAI's existing large language models, we have developed a new class of recommender algorithms that can be instructed through natural language. We also have built-in mechanisms to periodically poll users on their wellbeing and proactively give them the option to adjust their feed.
So, nothing? I can see this being generated and then reposted to TikTok, Meta, etc for likes and engagement.
It's insanely impressive. At the same time, all these videos all look terrible to me. Still get extreme uncanny valley and literally makes me sick to my stomach.
This stuff works really well when you make something that's exaggerated reality, as in either an animation or a MTV-style music video
I can't find the link now, but I saw a continuous shot video of a grocery store from the perspective of a fly. It was shot in the 90s music video style and looked so damn good.
Some of the stuff being done by these guys is also a whole lot of fun (slightly NSFW and political content), and it fits the music video theme:
https://www.youtube.com/watch?v=V4zwIhS2iZk
Agree - leaps and bounds beyond anything I would have dreamed possible a few years ago...but... IDK, if I'm honest, the sound was way off too, not just the visuals. The music sounded detuned slightly, and the crowd noise was "crackly" etc. etc. It had a low-fidelity "quality" to it.
Personally, I feel mixed feelings. I'm impressed, but I'm not looking forward to the new "movies" that are going to litter YouTube et al generated from this.
They seem like they're low FPS videos. I wonder if they're rendering 24 FPS and it's mismatching youtube's 30 FPS and causing the weird stuttering.
Maybe by Sora 3 they will have scene consistency. Gah it's so jarring to me that the poll the racing ducks are in just randomly changes. My brain can tell it's not consistent scene to scene and feels so jank.
Will something like Sora 2 actually be used in Hollywood productions? If so, what types of scenes?
I imagine it won’t necessarily be used in long scenes with subtle body language, etc involved. But maybe it’ll be used in other types of scenes?
I saw a famous actor-director (can't remember who, but an A-list guy) said it would be super valuable even if you only use it for establishing shots.
Like you have an exterior shot of a cabin, the surrounding environment, etc — all generated. Then you jump inside which can be shot on a traditional set in a studio.
Getting that establishing shot in real life might cost $30K to find a location, get the crew there, etc. Huge boon to indie films on a budget, but being able to endlessly tweak the shot is valuable even for productions that could afford to do it IRL.
Probably Ben Affleck. https://www.youtube.com/watch?v=ypURoMU3P3U
Searched around and found it. It was actually Ashton Kutcher's interview with Eric Schmidt.
Kutcher mentions the establishing shots, and I'd forgotten also points out the utility for relatively short stunt sequences.
> Why would you go out and shoot an establishing shot of a house in a television show when you could just create the establishing shot for $100? To go out and shoot it would cost you thousands of dollars.
> Action scenes of me jumping off of this building, you don’t have to have a stunt person go do it, you could just go do it [with AI].
Casey Affleck is currently shooting a horror vampire period piece using Comfy UI and an Unreal Engine Volume. The AI is used for the background plates. It's just a test, but it's happening right now.
Jason Blum is also getting really into the tech.
Wow. What an intelligent take. I would have never expected this from Ben Affleck. He seems extremely familiar with the technology and it's capabilities and limits.
People use these for sure but the biggest problem with these I feel is that they produce ”finished shots” with 8 bit colors and heavy grading. It’s hard to mix it with the other material which actually looks quite bland while it is being worked on. Would be great if somebody would train a model on raw footage.
I think it will mainly be used for inpainting and outpainting, i.e. for adding and removing stuff in a scene. Things that currently have to be done with relatively expensive CGI. More complex things, especially things that have to look identical between scenes, or have to look a very specific way, will still require filming or classical CGI, or both. (For now.)
Ugh. While technically extremely impressive, I'm so tired of the slop. Every AI content generation tool should have a watermarking system in place, and sites like YouTube should have a way to filter out AI generated content from search results with the press of a button.
Ever since the launch of Veo, there's already so much AI slop videos on YouTube that it becomes hard to find real videos sometimes.
I'm tired, boss.
I feel that this is a data collection activity (and thus, more advanced future models and usecases) disguised as a social media. People will provide feedback in the form of clicks/views on AI generated content (better version of RLHF) on unverified/subjective domains.
Biggest problem OpenAI has is not having an immense data backbone like Meta/Google/MSFT has. I think this is step in that direction -- create a data moat which in turn will help them make better models.
Lets take a step back and realise how incredible this is (I'm sure there are plenty of other `ackshually` comments)
Can it do Will Smith eating spaghetti? (I can't get access in UK)
Wish I was cool enough to have an invite code. Oh well, as an iOS build nerd next best thing I can do is inspect their ipa I guess. Interesting that they have some pretty big duplicate mp4s nobody caught in NoFaceDesignSystemBundle: cameo_onboarding_0.mp4 & create_ifu_1.mp4 | 7.3MB and cameo_onboarding_2.mp4 & create_ifu_0.mp4 | 5.2MB.
Also I find it neat that they still include an iOSMath bundle (in chatGPT too), makes me wonder how good their models really are at math.
in a computer graphics course i took, we looked through how popular film stories were tied to the technical achievements of that era. for example, toy story was an story born from the new found ability to render plastics effectively. similarly, the sora video seems to showcase a particular set of slow moving scenes (or when fast, disappearing into fluid water and clouds) which seem characteristic of this technology at the current moment in time
What is the target market for this? The videos are not good enough for YouTube. They are unrealistic, nauseating and dorky. Already now any YouTube video that contains a hint of "AI" attracts hundreds of scathing comments. People do not want this.
Let me guess, the ultimate market will be teenagers "creating" a Skibidi Toilet and cheap TikTok propaganda videos which promote Gazan ocean front properties.
Soon, you won't even have to do anything to post a video of yourself doing something "interesting" on social media, what at time to be alive.
There would for sure be large swathes of people who would just lie about what they're doing and use AI to make it seem like they're skateboarding, or skiing or whatever at a pro or semi-pro level and have a lot of people watch it.
That dragon flew backwards at one point didnt it.
Impressive that THAT was one of the issues to find, given where we were at the start of the year.
From watching the video I have an impression that these guys just want to appear cool, and the product looks like that too. To appear to be very cool, for people who won’t ever use it, apparently. Same impression I’ve got from watching that promo with Jony Ive. Beautiful, and don’t you dare to think it through.
Did they make human voices sound robotic on purpose? Is that some kind of Ai fingerprinting? It's way too obvious
It's very hard for simultaneous good audio generation with video generation (simultaneous generation is necessary to maintain lip sync). Veo 3 et al also have flat monochannel audio, but not as bad as these Sora 2 demos.
It's fitting that they host the video on Youtube, since that is where all of their training data came from.
This is super cool and fun and will almost certainly be really bad for society in loads of different ways. From the descriptions of all the guardrails they're needing to put in it seems like they know it too.
Glad to see someone is looking out for a forest, here. A diverse host of excuses have cropped up to explain away the anxiety AGI brings, and I totally understand why. Yet again, today we stare into the abyss.
This seems like a good time to remind ourselves of the original OpenAI charter: https://web.archive.org/web/20230714043611/https://openai.co...I wonder how exactly they reconcile the quote above with "We are concerned about late-stage AGI development becoming a competitive race without time for adequate safety precautions"...
I am not for or against AGI, but why is there anxiety around it? Do people simply hear sales rhetoric and assume that it can exist and will be used in order to dominate their lives?
I'm not referencing sales rhetoric, I'm referencing scientific consensus. AGI will have the same kind of impact on our species as fire and electricity did. We stand at a crossroads between unimaginable success and enormous catastrophe...
Well, good luck with that, hopefully it will learn to spell blueberry.
But think of all the 0 legitimate use cases for this technology.
"Please don't post shallow dismissals, especially of other people's work. A good critical comment teaches us something."
"Don't be snarky."
https://news.ycombinator.com/newsguidelines.html
tbh I didn't know it was technically possible for dang to be downvoted
it helps altman with world domination, so one legitimate use case for one person?
So being able to generate sign language videos for people who cannot hear is not a legitimate use case for AI videos? Or is your hate boner for AI just blinding you from useful applications?
Please don't respond to a bad comment by breaking the site guidelines yourself. That only makes things worse.
(Your comment would be just fine without the last sentence)
https://news.ycombinator.com/newsguidelines.html
Fair enough, I will keep that in mind.
Great point. Really, the main problem with subtitles is that the creator can understand them without having to know another language, and therefore can spot check them. That makes it much more difficult to insert Black Mirror-style Contextually Relevant Advertisements.
Is there a reason that's superior to subtitles, which are already fairly easy to generate?
sign languages are completely different languages from spoken languages, with their own grammar etc.
subtitles can work but it's basically a second language. perhaps comparable to many countries where people speak a dialect that's very different from the "standard" written language.
this is why you sometimes have sign language interpreters at events, rather than just captions.
there's not really a widely accepted written form of sign language.
> subtitles can work but it's basically a second language
That argument applies just as equally to sign language - most countries have their own idiosyncratic sign language. (ASL, LSE, etc.). Any televised event that has interpreters will be using the national language version.
The closest thing you're thinking of is IS - International Sign but its much more limited in terms of expression and not every deaf person knows it.
> there's not really a widely accepted written form of sign language.
Because it makes no sense to have it unless there was a regional deaf community that was fluent in sign language and also simultaneously illiterate.
https://www.reddit.com/r/NoStupidQuestions/comments/6t7k1w/h...
>this is why you sometimes have sign language interpreters at events, rather than just captions.
No, the reason is because a) it's in real time, and b) there's no screen to put the subtitles on. If it was possible to simply display subtitles on people's vision, that would be much more preferable, because writing is a form of communication more people are familiar with than sign language. For example, someone might not be deaf, but might still not be able to hear the audio, so a sign language interpreter would not help them at all, while closed captions would.
if you're maximizing accessibility you'd have both. often in broadcasts with closed captioning, there will also be a video of the sign language interpreter.
Sometimes the captions miss things or are really terribly written.
Why can’t sign language be written? Why does it need to be on video?
The real question is... what is the advantage of written sign language versus... normal writing? I think a lot of people are confused and think that there is only one universal form of sign language used worldwide [1].
Second problem is that sign language is heavily influenced with corresponding facial expressions, body language, the motion of the hands, even how emphatic the motions are. Trying to approximate what is effectively a SPATIAL language into written glyphs feels like a complete waste of time.
[1] https://en.wikipedia.org/wiki/List_of_sign_languages
> what is the advantage of written sign language versus... normal writing
If your native language is French, why might you prefer things to be written in French rather than, say, Swahili?
I feel like we might be talking past each other but it is funny that you chose French and Swahili. [1]
The point is that "SIGN LANGUAGE" is idiomatic to the native speaker's tongue. So if you're going to take the time to create a specialized written form of it, you can just write using the native language which can be read by BOTH the Deaf and non-Deaf community.
Deaf people are not magically illiterate.
Creating a written sign language serves no value since it is just a crappier version of the normal written equivalent.
So there's not a lot of value in creating a written form of say the French Sign Language because you can just use French.
Swahili regions have multiple types of sign language including Kenyan Sign Language.
[1]
https://en.wikipedia.org/wiki/French_Sign_Language
https://en.wikipedia.org/wiki/Kenyan_Sign_Language
> The point is that "SIGN LANGUAGE" is idiomatic to the native speaker's tongue.
No, this is not true. French and French Sign Language are totally unrelated languages. Sign languages generally have little to do with the spoken language of the country they’re used in, that’s why for example American Sign Language and British sign language are completely different and not mutually intelligible despite the UK and the US speaking the same language (with only slight differences in accent and vocabulary).
There isn't a standard written form of any major sign languages
Yes, but there’s no fundamental reason why there couldn’t be one. It’s not a good reason to accept all the downsides of AI.
AI is irrelevant to the reason why there isn't a written version of every single national dialect of sign language. The reason it doesn't exist is because it would serve no purpose (source: many deaf friends). Deaf communities learn the country's writing system just like everyone else.
The closest thing out there is SignWriting [1] which has about as much traction in the real world as esperanto.
https://en.wikipedia.org/wiki/SignWriting
ASL would be the target, but also the hand motions might not be universal to convey.
LOL. Yeah, that's way better than closed captions, even auto-generated ones.
THANK YOU. READING SOME OF THE COMMENTS IN THIS THREAD IS MAKING ME FEEL LIKE I'M TAKING CRAZY PILLS.
If you're going to convert audio to a digital form in realtime anyway we have this new amazing invention called the WRITTEN LANGUAGE.
Holy over-engineering batman! Is text too old-fashioned?
"download the Sora app"
click
takes me to the iPhone app store...
I think someone had called it many months back (and in fact I felt it too) that the feed for Sora seemed very much like a social media app. Then the only thing left was to make it into vertical scrolling with videos and voila you have your tiktok clone.
Why do you have to download an app to use Sora 2 (vs it being available on the web like ChatGPT)?
Prediction: we'll see at least one Sora-generated commercial at the Super Bowl this year.
Any idea if or when it will be available in EU? https://apps.apple.com/us/app/sora-by-openai/id6744034028
edit: as per usual it's not yet...
Amazing that even Sora2 can't make Sam Altman not look like a w@nker.
Shameless plug but I am creating a startup in this space called cleanvideo.cc to tackle some of the issues that will come with fake news videos. https://cleanvideo.cc
This is a good comparison thread of capabilities of sora vs sora 2
https://x.com/mattshumer_/status/1973085321928515783
While the quality of what I'm seeing is very nice for AI generated content (I still can't believe it) but the fact thay they are mostly showing short clips and not a long connected consistent video makes it less impressive.
Does this survive panning the camera away for 5 to 10 seconds and then back? Or basic conversation scene with the camera cutting between being located behind either speaker once every few seconds?
Basically proper working persistence of the scene.
Dude, this generation of AI video models are just starting to have basic camera production terms understood, and then it is exactly like LLM generation: it's a pull of a slot machine arm; you might get what you want, but that's "winning" and the slot machine only gives out winners one in every 100 pulls. Every possible thing that could not be right happens.
For example, I'm working with a walking and talking character at this time using multiple AI video models and systems. Generated clips any length longer than 8 seconds risk rapid quality loss, but sometimes you can get up to 12-19 seconds without the generation breaking down. That means one needs to simulate a multiple camera shoot on a stage, so you can cut around the character(s) and create a longer sequence. But now you need to have multiple views of the same location to place your character(s) into - and current AI models can't reliably give you a "different angled views" of an environment. We just got consistent different views of characters, and it'll be another period until environments can be generally examined from any view. BUT, that's if people realize this is not in the models yet, and so far people are so fascinated by the fantasy violence and sexual content they can make nobody realizes you cannot simply "look left and right" in any of these models and that even works with consistency or reliability. There are workarounds, like creating one's entire set and environments in 3D models, for use as the backgrounds and starting frames, but that's now 3D media production + AI, and none of the AI tools generate media that even has alpha channels, and a lot of similar incompatibilities like that.
OpenAI launches Sora 2 in a consumer app to collect RL feedback en masse and improve their world models further.
Their ultimate goal is physical AGI, although it wouldn’t hurt them if the social network takes off as well.
Find this sort of innovation far less interesting or exciting than the text & speech work, but it seems to be a primary driver of adoption for the median person in a way that text capability simply is not.
Video generation is extremely exciting a.k.a. https://video-zero-shot.github.io/
However, personalization (teleporting yourself into a video scene) is boring to me. At its core, it doesn't generate new experience to me. My experience is not defined by photos / videos I took on a trip.
I also can't think of a reason why I would ever want to look at an AI generated video.
however as they hint at a little in the announcement, if video generation becomes good enough at simulating physics and environments realistically, that's very interesting for robotics.
A little tangential to this announcement, but is anyone aware of any clean/ethical models for AI video or image generation (i.e. not trained on copyright work?) that are available publicly?
Sora 2 is a lot of fun. Using it feels like a glimpse into the future. The last time I felt this was with the Sesame voice demo.
That sure seems to be getting close to something usable for movies...kinda.
Sam looks weirdly like Cillian Murphy in Oppenheimer in some shots. I wonder whether there was dataset bleedover from that.
If I was on the OpenAI marketing team I maybe wouldn't have included the phrase "and letting your friends cast you in their [videos]". It's a little chilling.
The livestream showed an interesting UX with Facebook-style permissions that make it so you very explicitly have to opt into this feature: https://bsky.app/profile/minimaxir.bsky.social/post/3m22zg2h...
Even moreso than Facebook tags, the person being cast can cause the deletion of the source video at any time.
> this app is not available in your country or region
The fact that no one talking about how it compares against Veo tells me everything I need to know. This page is now filled with some bots!
The model's quality is incredible, but more tools are needed to take advantage of its capabilities, this is kinda the magic of open models.
How did they generate the videos with Sam Altman. Did they just provide a picture of his face and then use him in their prompts?
You can use the "cameo" feature only with users who have gone through the cameo creation flow. Sama has an account and created a cameo likeness of himself. When you create your cameo you can choose who is allowed to make videos using it: "only me", "people I approve", "mutuals", or "everyone".
TTS is horrible compared to Google's veo 3
I hate this vacant technology tbh. Every video feels like distilled advert mindless slop.
There's still something off about the movements, faces and eyes. Gollum features.
So far the true progress it has made is getting textures right close up. It still fudges how skin looks like the more it pans away from the characters.
These models will churn out shitty advertisements, just give it a few years at most.
I really hope they have more granular APIs around this.
One use case I'm really excited about is simply making animated sprites and rotational transformations of artwork using these videogen models, but unlike with local open models, they never seem to expose things like depth estimation output heads, aspect ratio alteration, or other things that would actually make these useful tools beyond shortform content generation.
hi there! would love an invite code if anyone reading this has a spare. really interesting stuff— thank you in advance! email is nick@usmobile.com
How far out are we from doing this in real time? What’s the processing/rendering time per frame?
https://deepmind.google/discover/blog/genie-3-a-new-frontier...
could already do it in real time by dimming the lightbulbs of a city or two.
Entering code 123456 reveals Sora 2 is only available in US/Canada region.
I just had a thought: (spoilers Expanse and Hyperion and Fire Upon the Deep)
Multiple sci-fi-fantasy tales have been written about technology getting so out of control, either through its own doing or by abuse by a malevolent controller, that society must sever itself from that technology very intentionally and permanently.
I think the idea of AGI and transhumanism is that moment for society. I think it's hard to put the genie back in the bottle because multiple adversarial powers are racing to be more powerful than the rest, but maybe the best thing for society would be if every tensor chip disintegrated the moment they came into existence.
I don't see how society is better when everyone can run their own gooner simulation and share it with videos made of their high school classmates. Or how we'll benefit from being unable to trust any photo or video we see without trusting who sends it to you, and even then doubting its veracity. Not being able to hear your spouse's voice on the phone without checking the post-quantum digital signature of their transmission for authenticity.
Society is heading to a less stable, less certain moment than any point in its history, and it is happening within our lifetime.
Can this be used to make hyper-realistic video games, or it's not that real-time yet?
Why is the video player so laggy?
Right? It constantly dropped frames for me (Firefox/Android).
VFX artists are definitely feeling the AGI / considering other career paths today.
VFX artist and developer here, who's deep into this stuff, and it is really not there. It's an island of itself, barely controllable and barely usable with other media. They are just now getting around to generating alpha channels, with virtual none of the existing pipelines for any AI video or image generation tools to even incorporate and work with alpha channels. This is just one of several hundred aspects of incompatibility. It really seriously appears as of no one at any of the AI video generation research teams has any professional media production experience, or even bothered too look at existing media production data standards, and what they are making tool-wise is incompatible in every possible respect.
"It really seriously appears as of no one at any of the AI video generation research teams has any professional media production experience, or even bothered too look at existing media production data standards,"
I had to chuckle at this. Because the arrogance of OAI et al will finally get them in the end when these projects continue to be negative NPV.
Honest q - do you think these things will make a big difference if these videos can be made in 15 minutes for $20 or whatever?
Won’t the industry change to adopt that massive price cut/productivity gain?
The cost is and will be more than that, the time will be more, and I really think people are underestimating the time it takes to create good stories. Sure, there will be online locations to make short form video of all kinds. People have had video cameras in their pockets for a very long time and being hobby film makers are not really popular. The AI video sites now are 95% people fascinated with the ability to make video at all, and after a bit their interest dies because to actually make anything that requires real work even with AI helping left and right. Consistency is a harsh mistress; and AI video is only good with it for a short duration. So any narrative that makes a story worth watching, it's not AI slop, will continue to require humans and human creativity - for the consistency that gives a story the integrity that makes it worth watching. At least for audiences that care. No doubt, there are commercial forces working to develop audiences that like and prefer AI slop.
Do you even see a path from the current AI systems to something that has that near-total control over every detail that is required for high quality VFX work?
Yes. Adding alpha channels would be step one. Then perhaps incorporate the "element" concept that is basically any identifiable visual anything; which is what VFX uses as a composite-capable element. Then build a whole visual scene description prose that is what we give to a video AI, and that prose is high level language where necessary and element-wise specific where necessary. Base that scene description prose on the language used by film makers directly, just adopt their terminology, and then track the industry's jargon within the models. That way anyone working in media will auto-magically know how to control them.
We are at a point now where it is now how to write software that is the problem but how to describe to the software that is the problem. Video and film making is so generalized, AI needs more information. Typically that information comes from a director's and their team's consistency during production. AI has neither the information for consistency of imagery nor the narrative and the perspective of the narrative a human director and team bring. In time, AI will develop large enough contexts, but will the hardware to run that be affordable? There is a huge amount of context in both an entire script and the world view perspective a film crew brings to any script, and for that reason I think many of the traditional (VFX included) film roles are not going to suddenly disappear. AI video does not replace their consistency at their budget, hands down.
When AI video is able to be just a part of the skill set, for example when it is compatible with compositing, editing, and knows that terminology, AI video will be adopted more. Right now, it is designed as an all or nothing offering.
Not with the way this thing renders hair (or any other high fidelity texture) https://x.com/GabrielPeterss4/status/1973090475486879818
I genuinely don't understand the consistent rhetoric on this site of:
> new AI feature/model comes out
> "it's going to replace people in this field! they better start looking for a new job!!!"
why is this a good thing?
It’s not a good thing, but it’s definitely a thing. Most of us here on HN are going to be affected by this.
How many more years do you think you'll need to keep saying this before it's actually true?
New grads are already having a tough time. My own expectation is that every recession or downturn from here on out, there will be the typical rounds of layoffs but without the typical increase in hiring afterwards. Maybe no “we replaced you with AI” moment, more of a “no new hiring” tendency.
Who said it was a good thing?
Is this AGI in the room with us now?
We are just heading for Lovely All TM.
I kid.
Art should require effort. And by that I mean effort on the part of the artist. Not environmental damage. I am SO tired of non tech friends SWOONING me with some song they made in 0.3 seconds. I tell them, sarcastically, that I am indeed very impressed with their endeavors.
I know many people will disagree with me here, but I would be heart broken if it turned out someone like Nick Cave was AI generated.
And of course this goes into a philosophical debate. What does it matter if it was generated by AI?
And that's where we are heading. But for me I feel effort is required, where we are going means close to 0 effort required. Someone here said that just raises the bar for good movies. I say that mostly means we will get 1 billion movies. Most are "free" to produce and displaces the 0.0001% human made/good stuff. I dunno. Whoever had the PR machine on point got the blockbuster. Not weird, since the studio tried 300 000 000 of them at the same time.
Who the fuck wants that?
I feel like that ship in Wall-E. Let's invest in slurpies.
Anyway; AI is here and all of that, we are all embracing it. Will be interesting to see how all this ends once the fallout lands.
Sorry for a comment that feels all over the place; on the tram :)
"if it's not worth [writing/playing/painting...], it's not worth [reading/listening/looking...]"
I had a friend over for my last birthday before going to a venue. He had a huge framed painting he had made. It made me cry.
A prompt delivered by Amazon drones would obviously not be the same lovely moment.
So yes, I agree.
Cool - now let's see how much it costs in compute to generate a single clip. (Also, notice how no individual scene is longer than a handful of seconds?)
The AI generated Sam Altman doesn't look even vaguely human.
Can it generate an analog clock displaying a given time?
Even if it can't, that wouldn't make this demo any less impressive.
12,000+ "AI startups" have been obliterated.
CEO of Loopt makes a cameo at 1:28 in the youtube vid.
Can I finally redo the Star Wars sequels with this? :)
Didn't Star Wars end in 2005?
Yeah, we've "plateaued" all right.
Everyone has the widest eyes in these Sora videos.
This is going to be a disaster. We are never going to be able to trust a video again and in short order propagandists are going to be using this to generate god knows what.
Tens of billions in funding and they've just built a modern version of JibJab[1]. Can't wait to start receiving this in reply-all family emails.
[1] https://youtu.be/z8Q-sRdV7SY?si=NjuyzL1zzq6IWPAe
Honest question: What problem does this solve?
Thing that was previously very expensive, manual and took a long time to do, and is done A LOT, is now made faster and cheaper by computers.
Pretty much the same problem we all work on every day in $DAY_JOB.
Fun and games until someone uses a tool like this to scam your family
It's ok, they're making the market for anti-ai tools much much bigger. (whether those tools work or not is a different issue)
It facilitates the generation of political propaganda.
OpenAI needing something to show to investors to say "See, this is why we need $1T."
What problem does what solve? Video generation models in general or Sora 2 specifically?
Liberation of employers from the shackles of their employees
After using Wan with comfyui, im uninterested in closed platforms. they lack the amount of control even if the quality might be better.
Instagram reels are gonna get crazy
You see the one with the dolphin on the trampoline?
Those `nature is amazing type of videos` are already flooded with AI
I've got used to immediately checking availability. In this case - iPhone app is US + Canada only and the website is invite only.
Going back to sleep. Wake me up when it's available to me.
No android app right?
Sora 2: Sloppy Seconds
This is the kind of thing people get excited about for the first couple of months and then barely use it going forward. It's amazing how quickly the novelty of this amazing technology wears off. You realize how necessary meaning/identity/narrative is to media and how empty it gets (regardless of the output) when those elements are missing.
Nicely cherry-picked.
Matrix here we come!
full-on productisation effort -> no AGI in sight
So, this is the AI Slop generator for the AI SlipSlop that Altman has announced lately.
Brave new internet, where humans are not needed for any "social" media anymore, AI will generate slop for bots without any human interaction in an endless cycle.
Only iphone...
Sora 2: Frato
AI Sam Altman is terrifying, holy shit. Squarely in uncanny valley for me.
Came here to say this myself. Would like to unsee that.
hey @simoncion looks like they are doing this for self-promotion that's against the site's guidelines
Now videos will be generated on the fly based on your preference. You will never put your phone down, it will detect when your sad or happy and generate videos accordingly
I wonder how this will affect the large cinema production companies (Disney, WB, Universal, Sony, Paramount, 20th century...). The global film market share was estimated to be 100B in 2023. If the production cost of high FX movies like Avengers Infinity War goes down from 300M$ to just 10K$ in a couple of years, will companies like Disney restrain themselves to just release a few epic movies per year? Or will we be flooded with tons of slop? If this kind of AI content keeps getting better, how will movies sustain our attention and feel 'special'? Will people not care if an actor is AI or real?
If anyone is feeling generous with one of their four invite codes, I'd really appreciate it. I'm at adrian@apetre.sc.
Interesting that they're going with a "copyright opt-out": https://www.reuters.com/technology/openais-new-sora-video-ge...
I guess copyright is pretty much dead now that the economy relies on violating it. Too bad those of us not invested into AI still won't be able to freely trade data as we please....
This makes me less excited about the future of video, not more.
It's technically impressive, but all so very soulless.
When everything fake feels real, will everything real feel fake?
Truly wonder if there will be some kind of renaissance in the video making domain when all settles down and this becomes the new normal.
Tastes and preferences are dynamic. It will certainly happen.
The ease of creating visually titillating media, coupled with the difficultly of consistency works against the creation of narrative media. I sure hope we don't get a generation of non-narrative beautiful slop.
it should be renamed into sore ai
We're cooked.
I think HN is too political like this tech is clearly amazing and it’s great they shipped it there should be more props even if it’s a billion dollar company.
There's a great lyric from ELUCID I think about when people say stuff like this:
> I don't have the privilege to think everything ain't political
i guess what I am saying is, though everything is political, it doesn't ahve to be "so" political.
Yeah I suppose so, most comments on this kinda thing are not really discussing the technology in a vacuum. I imagine it's due to the quite cynical nature of HN at this particular time period where society is fundamentally shifting, in arguably a negative direction, with this kind of technology as one of the main reasons. I haven't been on HN for that long, what was it like 5-10 years ago? I'm curious how it will be in 5-10 years.
Yes the tech is amazing. But tech is not everything, after 20 years of social media, its pretty clear to everyone that those things can have large long term impact both positive and negative for society, discussing the potential impacts of the tech is not being "political", its just being interested in the future.
Im not sure, what if society only learns through hardships?
Well certainly if you don’t want us to discuss the possible implications ahead, then yes we can only close our eyes and learn from hardship once it’s there, but then what do we learn ? To not close our eyes next time ? We could just do that like now.
I think it depends on the time and place. So in this context it doesnt matter too much, its a billion dollar company, but if it was a guy with a project he just spent all year on, publishing it on HN for the first time, I would expect people to focus less on teh politics and more on the achievement which is something I dont see too often, unless the work is stellar. Perhaps there is just a high bar on HN
Sam still pretending they’re close to AGI in the trailer lmao
impressive engineering that's hard to see as a net good for humanity.
it doesn't spark optimism or joy about the future of engaging with the internet & content which was already at a low point.
old is gold, even more so
Anyone have an invite they want to share with me lol.
What is the point? Who wants to watch these videos?
“With Sora 2, we are jumping straight to what we think may be the GPT‑3.5 moment for video.”
I think feeling like you need to use that in marketing copy is a pretty good clue in itself both that its not, and that you don’t believe it is so much as desperately wish it would be.
I am looking at the videos and really had a feeling that it looks right (minus a lot of obvious fuck ups still) where previously something felt fundamentally wrong with ai videos. It feels somewhat important, in so far you consider ai generated videos important.
It's the skin textures. It's the slightly better lipsyncing. Maybe it will be different when us normal users get it but so far the demos with Sam don't make him look waxy.
The Sora app squaring off against Meta's social video app is the real story here.
Sora 2 itself looks and sounds a little poorer than Google Veo 3. (Which is itself not currently ranked as the top video model. The Chinese models are dominating.)
I think Google, with their massive YouTube data set, is ultimately going to win this game. They have all the data and infrastructure in the world to build best-in-class video models, and they're just getting started.
The social battle will be something completely different, though. And that's something that I think OpenAI stands a good chance at winning.
Edit: Most companies that are confident of their image or video models stealthily launch it on the Model Arena a week ahead of the public model release. OpenAI did not arrange to do that for Sora 2.
Nano Banana, Seedream/Seedance, Kling, and several other models have followed this pattern of "stealth ELO ranking, then reveal pole position".
https://artificialanalysis.ai/text-to-video/arena?tab=leader...
The fact that this model is about "friends" and "social" implies that this is an underpowered model. You probably saw a cherry picked highlight reel with a large VRAM context, but the actual consumer product will be engineered for efficiency. Built to sustain a high volume of cheap generations, not expensive high quality ones. A product built to face off against Meta. That model compete on the basis of putting you into videos with Pikachu, Mario, and Goku.
> I think Google, with their massive YouTube data set, is ultimately going to win this game.
I don't know, applying the same thinking to LLMs, Google should have been first and best with just text based LLMs too, considering the datasets they sit on (and researchers, among others the people who came up with attention). But OpenAI somehow beat them on that regardless.
The problem for Google existed with the infobox at the top of search results. If users get the answer to their query without having to visit the web page where the answer came from, and where Google shows the ads, means that users don't see ads, and that website operators don't get ad revenue. ChatGPT was Google's Kodak digital camera moment. They had internal transformers-based chatbots (that really wanted to send you pizza, for some reason), but deploying that would have cannibalized their existing business model, so in the meanwhile, their lunch got eaten by an outside competitor.
"Thou shalt not create a machine in the likeness of a human mind."
Ah, a holy scripture from the Orange Catholic Bible!
As usual: impressive until you look close. Just freeze the frame and you see all the typical slop errors: pretty much any kind of writing is a garbled mess (look at the camera in the beginning). The horn of the unicorn sits on the bridle. The buttons on Sam's circus uniform hover in the air. There are candleholders with somehow candles inside as well as on top. The miniature instruments often make no sense. The conductor has 4 fingers on one hand and 5 on the other. The cheers of the audience is basically brown noise. Nedless to say, if you freeze the audience, hands are literally all over the place. Of course, everything conveniently has a ton of motion blur so you cannot see any detail.
I know, I know. Most people don't care. How exciting.
Is your complaint that it has errors? I mean look at what it can do. This is a freaking computer generating things from scratch based on a prompt. Two years ago, technology like this was so much worse and could only generate basic images and videos. Now it can generate visuals all from the text someone puts in.
Anyone, literally anyone, can use it (eventually) to generate incredible scenes. Imagine the person who comes up with a short film about an epic battle between griffins and aliens...Or a simple story of a boy walking in the woods with their dog...Or a story of a first kiss. Previously people were limited to what they had at hand. They couldn't produce a video because it was too costly. Now they can craft a video to meet their vision.
I do find it exciting.
> Is your complaint that it has errors?
Well, yes? There's a reason why everything that was produced with these tools so far is garbage: because no one actually caring about their art would accept these things. Art is a deliberate thing, it takes effort. These tools are fine for company training videos and TikToks. Of course a few years ago this was science fiction. They are immensely impressive from a technical perspective. Two things can be true.
There is that magic word again, “eventually”. When is that? The same time we get warp drives?
I don't know if it's just me or other people are feeling it as well. I don't enjoy videos anymore (unless live sports). I don't enjoy reading on my monitor anymore, I have been going back to physical books more often. I am in my early thirties.
The point is that sora2 demo videos seemed impressive but I just didn't feel any real excitement. I am not sure who this is really helping.
Personally I can't wait for super creative and novel indie film productions as film production will be more liberated from the grip of Hollywood and the influence of the upper classes in general. Especially once the Chinese make less-censored-to-Western-users models more available and even more so once people can run these things at home in some years.
that sounds like clinical depresion, I'd check with my endocrinologist to get blood work done
Same with me !
So a social network that's 100% your friends doing silly AI things?
I feel like this is the ultimate extension of "it feels like my feed is just the artificial version of what's happening my friends and doesn't really tell me anything about how they're actually faring."
Social media also tends to highlight the best parts of people’s lives, creating unrealistic expectations and views for those consuming it and looking at their real life. Now social media won’t even be a highlight reel, but completely fabricated.
I have to imagine there will be a rebellion against all of this at some point, when people simply can’t take the false realities anymore. What is the alternative? Ready Player One? The Matrix? Wall-E?
Which seem to level the play field, at least virtually
Maybe inside of a social network specially for AI, but a concerning number of people don't realize images and videos are AI, even when it's bad AI. As it gets better, and starts integrating the poster's image (like Sora 2), that's going to get even worse.
Some people use filters/photoshop to artificially juice their images; now they can use AI to artificially juice every aspect of their on-line presence.
I just watched the announcement video and something about it just gives me the ick. The whole time I just had the uncanny valley feeling.
The technology itself is super impressive, but a social media app of AI slop doesn't feel like the best use of it. I'm old enough to not really be interested in social media in general anymore, so maybe I'm just out of touch, but I just can't see this catching on. It feels like the type of thing that people will download, use a few times until the novelty wears off and then never open again.
Sounds like a way to make everybody aware of Sora and its capabilities.
I bet the real goal is to make money from long tail of corporate market ( ads, info videos etc).
I built an MVP of this [1] with images (not video) and in more of an Instagram style (not tiktok) back in ‘22, with the tagline ‘What if social media were literally fake?’
I am bullish on this, albeit with major concerns in many domains. It was fun and addictive as hell with images. With video it will be wild.
[1] https://hardwork.party/cheese/
I find it comical that OpenAI with all the power of CharGPT even them are unable to release an app for both iOS and Android at the same time. Wow, good marketing for Codex.
That is more of a statement of the complete dominance of iPhones among gen z.
Or Sama's documented reverence for Apple products. We are talking about the guy who sold Tim Cook his AI for $0.00, he's not exactly got the horse drawing the cart here.
Google sold Tim Cook their search engine for $-25B per year
Not even for all regions for iOS
I’m eagerly awaiting for some unexpected social problems this crops up
Overall, appears rather underwhelming. Long way to go still for video generation. Also, launching this as a social app seems like yet another desperate try to productize and monetize their tech, but this is the position big VC money forces you into.
I've perhaps been away from the scene for a bit, but I'm very impressed. To me this is absolutely "video generation", and I don't get your disdain for productization and monetization; last I checked this wasn't "Basic Research News".
I don't think it's disdainful to point out the lack of PMF for a dedicated app for Sora, nor how its behind competitors who don't require a dedicated social app. No need to strawman the guy, I think it's okay to be reasonably critical of ideas still on this website.
Inb4 make your own video model and see how easy it is
Every AI video demonstration is always about funny stuff and fancy situations. We never see videos on art, history, literature, poetry, religion (imagine building a video about the moment Jesus was born) ... ducks in a race !? Come on ...
So much visual power, yet so little soul power. We are dying.
What do you imagine a generated video about poetry would be?
>Every AI video demonstration is always about funny stuff and fancy situations.
The thing about AI slop is that by its very nature, unless it's heavily reined in by a human, it's invariably lowest common denominator garbage. It very likely will generate something you yourself could think of within the first five seconds of hearing the prompt, not some very clever take on it, so it can only work as a placeholder (AI as a replacement of stock images is great, for example) or to add background detail where it won't call attention to itself and its genericity.
>imagine building a video about the moment Jesus was born
Given there are multiple paintings on the subject, I very much doubt no one has generated something like that already.
[dead]
[dead]
More discussion: https://news.ycombinator.com/item?id=45428122
Comments moved thither. Thanks!
Edit: looks like this post was actually first, so maybe we'll reverse the merge
[dead]
[flagged]
OpenAI is cooked.
Absolutely cooked.
After the disaster that was chatGPT4.001, study mode and now this: an impossibly expensive to maintain AI video slop copyright violater, their releases are uninspired and bland, and smelling of desperation.
Making me giddy for their imminent collapse.
[dead]
[dead]
Someone remind me the benefits of mass produced fake videos again?
People are doing cool things with it. Here’s one example:
https://www.tiktok.com/@dreamrelicc
Before AI, each video on this channel would have taken a large team with a Hollywood budget to create. In a few more years, one person may be able to turn their creative vision into a full-length movie.
I speak for everyone when I say we don't need these videos at all and would be better off without them
I disagree, so not everyone, I guess!
This is absolutely horrible.
People need to be exposed to what is real. Not more artificial stuff.
I think this is the point at which humanity will finally puke and reject this crap.
Just because a small segment of people like it doesnt mean the mass majority will.
I personally love Monet, he's not for everyone, I know, but I'm sure you can find some art you appreciate
You probably don't personally love AI generated impressionist content.
No, but there's some stuff that are really creative. Ironically I think the reason I'm more positive about it is because I only encounter AI generated (non-text media) ~ once a week / 2 weeks.
But modern AI could create images which are basically indistinguishable from a real Monet if you are not an expert. So the fact that you like Monet's pictures, but not Monet-like AI pictures, shows that part of what you like is the fact that an image is made by a specific human instead of being generated by a diffusion model.
I dunno, look at these [0] I think they're quite nice! But I can imagine getting bombarded with them all day can eventually turn someone off. (I assume I'd feel the same if I saw 100 Monet's every day for 3 years)
[0] https://andymasley.substack.com/p/a-ton-of-ai-images-ive-mad...
Maybe your real is good. For most people on Earth, real isn't that great.
What are the benefits of those videos?
What are the benefits of producing any video?
What are the benefits of this comment?
Challenging the value of AI generated "art"
Then the purpose of those videos is to challenge the value of non AI generated "art"
(half sarcastic, but you could make the argument that most art has no benefit besides to the person that made the art)
Nice! I enjoyed this sub thread. I'm not sure what I conclude but I enjoyed thinking about this.
> In a few more years, one person may be able to turn their creative vision into a full-length movie.
Yes, but at the same time the value of video production will quickly drop to 0. Or to whatever it costs to generate that video in terms of tokens.
The value will shift to search or curation - if the cost to produce drops to nil, then the value will be in finding good content amongst a flood of sameness.
Those videos look like some teenager thoughtlessly applying an aftereffects filter(whatever) to 1000 short selfie videos. On What planet would this require a Hollywood budget and years? Who are you shilling for exactly? Do you really believe what you write.
> People are doing cool things with it
Things are cool because they are unique, very hard to create, and require creativity. When those things become cheap commodities, they are no longer cool.
Exactly. Pushing a photo through a Van Gogh filter doesn't get near what a real Van Gogh expresses. It's in a temporal context, communicates something about the person and their thoughts about reality. Their artistic choices matter, because they can't just put out 10 different variations, instead they have to pick one. And then we can think about why that one was chosen.
The same could be said about software, and it's safe to say that open-source software making complex workflows easier and more efficient is a net good.
Making better tools is better for everyone: the median usage of those tools downstream is a separate issue.
If you're comparing how art is evaluated to how software is evaluated then it sounds like you only understand one or the other.
Indeed. Art is partially evaluated by how impressive it is. That's why posting AI images on social media won't yield a lot of likes anymore. People have gotten used to images being easy to create, so they aren't seen as valuable anymore. The same will be true for videos.
AI pictures today are much less impressive than Dall-E 2 pictures were a few years ago, despite the fact that the models are much better nowadays. Currently AI videos can still be impressive, but this will quickly become a thing of the past.
Then people will move from trying to create art to creating "content". That is, non-artistic slop. Advertisements. Porn. Meme jokes. Click bait. Rage bait. Propaganda. Etc.
I would argue that we just get pickier and more sensitive to slop. When everyone can make a movie, the standard for a good movie will be higher. Many current Hollywood films wouldn’t make the cut. But maybe some kid in Nigeria makes the greatest film of all time.
By that logic, some kid in Nigeria could have written the greatest book of all time. At least by commonly accepted measures, that didn't happen.
Hard to interpret that comment as anything but racist. Chinua Achebe is widely considered one of the greatest modern novelists. He was 28 when he wrote Things Fall Apart.
Perhaps learn the meaning of the phrase "by commonly accepted measures" before you accuse someone of racism. I'm pretty sure hardly anyone knows about Chinua Achebe, so your definition of "widely" must be quite wide.
Things Fall Apart has sold over 20 million copies and has been translated into more than 50 languages. It is a staple of literature curriculums in schools and universities across the globe. That isn't a "wide" definition of widely known; it's the standard one.
Then you have Chimamanda Adichie, who has sold millions of copies and won several awards, including the BBC National Short Story Award, widely described as "one of the most prestigious awards for a single short story"
Then another Nigerian writer, Wole Soyinka, won the Nobel fucking Prize in Literature in 1986. Or is that measure not good enough for you, your highness ?
Not only do you come across as racist, you clearly have no idea what you're talking about. Congratulations.
These examples seem highly cherry-picked. If you look at bestseller lists, or writers who average people actually know, the results are in fact very different. Your accusation ("racist") is defamatory.
Calling a Nobel Prize winner, among others 'cherry-picked' in an argument about literary greats where you asked for 'commonly accepted measures' is one of the most intellectually dishonest things I've ever read, so congratulations again.
You were thoroughly proven wrong so now your new standard for literary greatness is "writers that average people know" ? (which is really just code for 'writers I know', because millions do know those writers, I wasn't sharing some secret). I guess that means we can throw out Faulkner, Joyce, and Woolf in favor of whoever's currently at the top of the airport bookstore list.
It's not "defamatory" to point out that your argument, which began with a dismissive generalization about an entire country, was based on profound ignorance (the kind that wouldn't have taken anything more than a basic google search to remedy). You were corrected with facts. Instead of going, 'I stand corrected, sorry', you're doubling down. It just makes you look worse, and stupid.
This is the most basic racist playbook happening in real time, and you're the star. If you genuinely think you aren't then you need to take a long, good look at yourself.
I love the aesthetic in this person's videos, I just wish it wasn't on tiktok :(
The problem is, it isn't their aesthetic, it's a resynthesis of the aesthetic of someone else's work.
I wasn't claiming they had ownership of that aesthetic, and that sort of gets into philosophical questions about whether one can own such a thing anyways. I like the style and I'm glad they have the tools to bring it into clarity from the abstract.
This is terrible
It's fun: maybe not for everyone, but there's clearly sufficient interest in it.
Whether said fun is "worth" the social and economic costs is a separate issue.
Democracy? Strengthened! Nothing says “informed electorate” like not knowing if a politician actually said they support nazism or if it was just a hyper-realistic AI puppet.
Trust in media? Soaring! Why believe your eyes or ears when you can doubt everything equally?
Journalism? Thriving! Reporters now get to spend their days playing forensic video detective instead of, you know, reporting news.
Social harmony? Better than ever! Nothing brings people together like shared paranoia and the collective shrug of “I guess truth is dead now.”
Honestly, what could possibly go wrong?
lol i wonder if this will create a market for PKI at the image sensor level so that videos will be cryptographically signed and baked into the actual video stream with steganography.
Advertising: you (her) wearing new clothing before purchase, hair/glasses/makeup, make overs; guys after 3 months of gym membership, you driving the new car, you in this specific new home... etc, etc... I'm surprised this is not already everywhere, but people are too occupied making nsfw and fantasy violence clips.
Targeted advertising has become just manipulation. I don't know if personalized advertisement videos for everyone promoting a fake world that doesn't exist is really a benefit for the world...
If course it's not a benefit, but it's an advertising angle that will work very well with a class of gullible consumers, and that is enough to justify it being plastered everywhere. I don't write these rules, I just notice them.
I imagine it's incredibly useful for prototyping movies, tv, commercials before going to the final version. CGI will probably get way cheaper too with some hybrid approach.
Obviously this will get used for a lot of evil or bad as well
can you imagine a billion dollar company promoting their new pre-vis app?
I feel like that's missing the point of pre-vis anyway, its purpose is to lay down key details with precision but without regard for fidelity (e.g. https://youtu.be/KMMeHPGV5VE), a system with high fidelity but very loose control is the exact opposite of what they want.
Fun.
... how dare you, sir. That is entirely unacceptable and you will be reported to the ministry of proper living!
Regardless of the slop, some people will learn to use it well. You have stuff like NeuralViz - quite the sight! - and other creators will follow suit, and figure out how to use the new tools to produce content that's worth engaging with. Bigfoot vlogs and dinosaur chase scenes, all that stuff is mostly just fun.
People like to play. Let them play. This stuff looks fun, and beats Sora 1 by a long shot.
Hopefully it catalyzes
I can have an idea and see a video of something like my idea pretty quickly.
What are the benefits of what you do? Does anyone know?
- Political propaganda
- Scamming people at scale
- Nonconsensual pornography
- Juicing engagement metrics for fading social media sites
- The ongoing destruction of truth as a concept in our increasingly atomized and divided world
I think the last one takes the cake.
[dead]
[flagged]
already happening: https://x.com/ken_wheeler/status/1954343731579994593
its well underway already
[flagged]
Could you please stop posting unsubstantive comments and flamebait? You've unfortunately been doing it repeatedly. It's not what this site is for, and destroys what it is for.
If you wouldn't mind reviewing https://news.ycombinator.com/newsguidelines.html and taking the intended spirit of the site more to heart, we'd be grateful.
Show me a coherent video that lasts more than 5 seconds and was generated with the model and maybe I'll start to care.
It is very underwhelming. It seems like a step backward. Scam altman should be replaced before he runs the company to bankruptcy.
If Sora 2 is aiming for AI‑Tok, ScaryStories Live is the jump-scare cousin: real‑time POV horror from a photo + a sentence. No film school, no GPU farm—just “upload face, pick fear level, go.” It’s less cinema, more haunted mirror, and it ships in seconds. scarystories.live