This sort of thing keeps getting written. You will still be able to art, even if AI art becomes better. I can still, to this day, still go for a walk even though cars exist. Not because I need to get somewhere, necessarily, but for the love of walking.
This is only being litigated in this way I think because art people never thought automation would come their way, and they're thinking about this stuff for the first time. A silly comment from some GenAI exec is not much justification for such an article. People know that writing and making music is fun. Said exec is just paid to say the opposite.
>You will still be able to art, even if AI art becomes better.
I don't know, maybe all the artists, musicians, and poets will be too tired from doing whatever menial labor bullshit jobs are left post AI to create great art.
Why not direct our efforts into enabling more people to devote their lives to creating art instead of fewer? Wouldn't it be amazing if everyone was able to spend their time creating because the machines were doing the work we actually didn't want to do?
In the near future, we will have less art made by professionals and more art made by hobby artists (with or without AI support).
It will be a tough time for professional artists, but personally I don't believe we will "lose the human spark" in this transition.
If all AI art is soulless, new artists will rise that can put the missing part back in.
How many amazing potential creations have been lost because their creators did not have the time or money to make them? I'm looking forward to see what these people will create in the coming decade!
I think without financial support you're not going to get very good art. Some of the greatest artists of the renaissance were only able to spend all day painting because they had patrons. My city has lots of public art because the government spends money on it.
I believe there's a substantive difference between digital art tools like photoshop or illustrator and LLMs. LLMs do not create "art", they create images. Previously, you had to PAY an artist to create art. Now, you pay for an API to create an image loosely based on the specifications you give it. As a result, there is less art in the world.
Art models, to me, are very powerful paintbrushes. The technical skill of the artist is less important, but their ability to create art that makes people FEEL is more important than ever.
I think it's interesting how this always gets framed as reducing the number of creators.
I run an AI storytelling site with thousands of users and by most measures I wouldn't say it writes better than a practiced human, but what it does do is allow would-be consumers to become their own creators.
Some users spend $100+ dollars a month, not because it writes 10x better content than what a $10 Kindle unlimited subscription offers, but because even at 50% the quality they'd rather have the story they wanted to tell but didn't exist elsewhere.
For me AI art is not about replacing creators: it's about letting users go from passively consuming content made by people who have skills, time, and/or resources they lack to actively creating what they want to see in the world.
I'm not sure how compatible being against the democratization of creation really is with the narrative of defending the arts.
That’s it and with every invention the percentage stays the same 90% consume and 10% create. The insane amount that gets put into options of image models to create the maximum amount of images per hour is the dumbest approach possible for the technology at hand. Nobody is there that wants to consume it.
The reason why I'm personally convinced that generative AI for creative output will be bigger than any of the "PhD level agents" everyone is chasing is that gen AI can finally change "90% consume and 10% create", by changing what it means to consume.
People assume my storytelling tool is for writers because until relatively recently there was no other option: you couldn't make a writing tool for readers who only want to read, or a photography tool for viewers who only want to view, or a music tool for people who only want to listen.
But we're getting closer to tooling that makes it so frictionless to create, that you can insert the creation process in the consumption process, essentially turning it into curation.
-
Next week Gemini Flash comes out and we're going to have access to the first publicly available image model that can edit images with as much nuance and understanding as current SOTA LLMs do for text.
Everyone is focused on a possible plateau in intelligence, but steerability is the key for creative output, and there are so many angles to attack it from that we're at the infancy of what generative AI can do for creative output.
I recently got hit up out of nowhere by somoene who was like i'll pay you to storyboard a music video of a famous dead rapper. Think tupac hologram but somehow even worse in terms of digging someone up from the grave. I was like ok sure why not, I know this generative AI stuff. Got paid like a hundred bucks to start on this storyboard.
Thing is the story was so fucking stupid that I stopped after generating one image. They probably didn't make the story itself with AI but it sure was ridiculous, I basically quit as soon as they asked me to somehow include a bunch of jailed celebrities like diddy into the next storyboard image, getting out of it saying that it went against my legal capabilities to include real living people.
I hope you realize that some people with millions of dollars just have really head ass ideas they're going to put through your thing and things are going to get weirder sooner as a result. Like that idea would have ended at the drawing board with a bunch of people talking each other down from it not that long ago but now it's basically like that WKUK sketch about a grape out here [1]
IMO the real interesting questions your post brings up are:
- what kind of labor is the $100 your users are spending coming from
- where would these $100 have gone before your product existed
- where do these $100 go once they change hands from users to you
There are many answers, not a single answer, to these questions, but I think they're worth considering.
In a previous life I worked for a startup where, getting to know our users, I realized that if you broke it down the revenue was essentially money being funneled from hourly-wage workers to venture capitalists, which was a sobering realization.
Not one person in this world should have to motivate you to do the thing you love. This is a reckoning moment for a lot of people, artists included. Dig deep. Because the truth is that a lot of people do things, not for the love of and enjoyment of doing the thing in and of itself, but because of the satisfaction they get when they impress other people.
"You're so talented"
"I love your art"
"I wish I could do that"
And now there is that fear that the validation will go away. And what are you left with?
>Because the truth is that a lot of people do things, not for the love of and enjoyment of doing the thing in and of itself, but because of the satisfaction they get when they impress other people.
And there's nothing wrong with that. That's part of what makes art art. Creators are not solipsistically creating art for its own and their own sake: they do it for sharing it with other people, and getting the others' recognition and connection.
Some of the greatest rock music, for one, was written to impress the other sex - and some of the best dance music was written to see people dance and their bodies connect to it.
Machines already took almost all artist jobs a long time ago. Ever heard of the camera?
Machines already took almost all musician jobs a long time ago. Ever heard of recording?
The current winner-take-all art fields are not very healthy for humans to work in anyway, as there are infinity starving artists for every success. Eliminating the few remaining successes won’t really change much from the perspective of the average artist.
Almost all human art today is already done by hobbyists.
It's incredible to make the claim that almost all art is done by hobbyists considering how large the movie, tv and video game industries are by employee headcount.
Okay, compare that to the amount of hobbyists by headcount. Also subtract out the support roles in those industries that don’t constitute “artist” roles. Is a programmer an artist? Is someone working on color grading according to the specifications of someone else an artist? Subjective for sure, but at least for me most roles in those industries don’t qualify as “artists” since they don’t exercise substantial control over the art output.
As an illustrative comparison, if we apply the same “massive org chart” model used in film or video games to painting, each employee would apply a few brushstrokes according to requirements specified by a painting director. Is each of them an artist? Or is the painting director the artist?
> Why not direct our efforts into enabling more people to devote their lives to creating art instead of fewer? Wouldn't it be amazing if everyone was able to spend their time creating because the machines were doing the work we actually didn't want to do?
People are also doing that. Note you're not travelling thousands of miles on foot to tell me this message; nor are you and your family doing subsistence farming. And you probably have time to do art if you want, because of existing advances.
You have missed the point of his article. These things keep getting written, by smart people, BECAUSE other smart people like you are not getting that point.
Obviously, people can still produce art. What he’s talking about is the corrosion of the fabric of culture that will serve to discourage and coopt the art-making impulse. It’s like this: When I was working on the last chapters of my book, I spent a few weeks in a cabin with no internet. Why? Certainly, the mere presence of internet wasn’t interfering with my keyboard, right? Right. But part of my brain wanted to surf and play games online, and it was dragging the rest of my brain with it.
“Art” is a cultural dynamic, not just a matter of production. It is being colonized by people who use the word and don’t care about art. And some people who might have become great artists will now become frustrated consumers and nothing more— their energies dissipated.
I don’t know how big this problem really is. But that’s his concern. At least engage with it instead of implying that this Stanford professor is a moron who doesn’t understand how computers work.
Orlando springs to mind, the pavement just siddenly ends and people who see a couple walking with a kid pull over and offer assistance, from personal experience.
>when we work at our favorite activities — cooking, gaming, hiking, music-making, writing (to name a few) — the process is often not easy or pleasant or smooth.
This gets to the heart of what bothers me about using LLM to emulate Art. This is the justification people make for using it, but no Artists I know finds making Art consistently unpleasant. Sure, there are times you're blocked or tired but generally in the end you are satisfied or have leaned something helpful from the experience. This is all of life. This is why there is Art. If you just want a pleasant time, you're looking for entertainment, not creative endeavor. Just put on a VR headset and leave making Art to the real Artists.
Art is like anything. If you want to get good at it you need to discover how to get dopamine from it and exploit that to habit and mastery. This is how you get in shape, it's how you improve your golf game, how you develop software, raise kids. Yet for some reason humanity wants a cheat code for all of this. Pills to make you skinny, video golf to give you an advantage at the links, top-tier schools to make your kids awesome. We're hellbent on dumping the baby out with the bathwater because we think the point is putting the tub away.
AI songs present as a refined end-product but you don't have to treat them as anything more than killing time or B-grade inspiration for something else. Instrument skills will surely be preserved somewhere due to nostalgia, it's the industry gatekeeping that is coming to an unharmonious end along with all the silly status games if AI songs ever take off.
For instance, why does an indy band need an AI-free backstory but a popstar can outsource all their imagination to the producer who may or may not use autotune or AI riffs?
Because the kind of people who listen to indie music are not thr same as those listening to pop music. Indie artists are valued for their originality or authenticity. Pop music is valued for the spectacle it puts on, not just the sound waves that come out of your earphones.
Personally I don't see any point in listening to AI generated music. It's not something I'll ever be interested in, certainly not something I'd be willing to spend money on
I have been moved by the rare piece of AI genre music. Admittedly, most of Gen AI that gets updooted is repulsive though. This emerging field is not for the faint hearted and has terrible curation chops.
Are there any examples of popular artists playing instruments regularly at shows? Not formerly popular artists.
Nearly all the concerts I've been to recently were rap artists and DJs . Beyonce doesn't place a huge emphasis on instruments in her show, nor do the KPop artists. Do people go see Taylor Swift for the guitar? Sometimes someone will bring instruments on for a few songs.
Original composition still happens with instruments, usually but not always.
I have been a musician at this point for 35+ years. The quote from the Suno CEO is beyond absurd.
I have spent quite a number of hours with MusicLM in AI test kitchen. To me, that is actually a new form of music synthesis. Synthesis in the context of FM, subtractive, etc.
I suspect MusicLM had some tracks removed from the training data because it use to make absolutely wildly creative psytrance clips and now it just doesn't.
Suno on the other hand to me has basically been a complete joke. The training data is just not wide enough to do anything interesting and new.
I think the only way to really use these tools creatively would be to produce your own trained model with AudioLM. The creative use then is in the musical output and not selling the model as yet another SaaS for $x.99 a month.
Of course, you could use Suno to create loops and sample material. The way midjourney is the greatest thing possible for digital collage in Krita/Photoshop.
GenAI art though on its own be it midjourney, stable diffusion, Suno, MusicLM, whatever already jumped the shark for me a year ago. The problem ultimately is the output is just so very limited.
> Suno on the other hand to me has basically been a complete joke. The training data is just not wide enough to do anything interesting and new.
Hard disagree, Suno is terrific, but likely depends on what you define as "new". Of course AI models can't explore outside the data they are trained on, but they can effectively generate within the parameter space, including areas unexplored directly by the content it was trained on.
Have done some really interesting mashups of styles that people aren't doing and the results are great imho.
If you are trying to make it do very specific things like you would another instrument, it's not the right tool.
I really enjoyed the perspective of this author and the article's contemplative tone. I think an important perspective is that these are tools for consumers, not creators. While we are all obviously both to a degree, the current economic model makes a clear distinction.
When it comes to monetary exchange, we are either artists or consumers. AI art has the potential to create a massive imbalance between these two fields, reducing the friction of being a consumer, and greatly reducing the number of people who are able to support themselves as artists. If you view yourself chiefly as a consumer this might (very narrowly) be viewed as a positive.
The social cost however is creating a society that favors consumption over creation, and weights incentives accordingly
> The social cost however is creating a society that favors consumption over creation, and weights incentives accordingly
We've been heading this way progressively for the last century.
If you look back at the early days of broadcast television, for example, there were only 3 channels. 72h of programming at most in any 24h time period. This deeply favored the creator (or perhaps more accurately, the owner of the channel of distribution). Now we have YT and Netflix, both of which are biased towards consumption. Is it all bad? In some ways; we have to wade through more to find the good stuff. But in other ways, we have more to choose from and more people can be creators.
A gigging musician I know, who is involved in many projects, including leading an on again, off again BiG BanD, recently had another aquaintence, play a recording to him of the soul song that this person had "written" with ai help.
I am struggling with comming to terms with the idea of artificial music.
Some people like it, so it will stay around,
and I cant begrudge them, at least for now, as ai will probably end up spawning new generas ,that would likely not come into bieng without it, which I suspect, the whole point will be inflicting there "art" on anyone they can.
This artificial art, has no back story, no sweat, no triumph, and will never have a voice or style it's own, it's only and explisitly derivitive, but there are those who are in some strange way comforted by that, a bieger shade of ecru, free from angst and any possibility of it showing anybody up.
Though all in all, I think it's, orrible, right nasty, and I imagine that it will end up banned
like a lot of other artificial things, that prove to be bad, for everybody they touch.
> AI music company Suno’s CEO was quoted as saying “I think the majority of people don’t enjoy the majority of the time they spend making music” in what sounds like yet another justification of Suno’s prompt-based (and very popular) generative AI music service...
> Such systems offer the promise of skipping the labor by bypassing the creative process and, yes, the difficulties, confusion, and frustration inherent in such endeavors, but…
The true purpose of AI is to free humans from creative and engaging work, so we can spend more time doing drudgery (like monitoring AI output).
> Statements like Suno’s perfectly captures the prevailing public mindset about AI: that Artificial Intelligence is little more than a labor-saving optimization tool. This mindset tends to be good for #Capitalism, but betrays not only a lack of understanding of why people make music, but also a profound lack of imagination regarding how we could, or would want to live with our technologies in our lives.
The true goal of Capitalism is the elimination of anything that isn't Capitalism, so it can be replaced with more Capitalism, until everything is Capitalism and nothing else remains.
Amusing to see a Stanford student complain about capitalism. Attending Stanford costs $70,000 a year. The bastardized John Adams quote: "I must study war so my sons can study business. My sons must study business so their sons can complain about capitalism."
Of all the billions of dollars in the world wasted on disturbing excess, aren't you thankful that some of it goes towards something worth more than what money can buy?
Can you explain your logic here? Why couldn't the student complain about capitalism while being enrolled at Stanford? I would have rather assumed that the high semester fee may perhaps contribute to them having that opinion.
I pay a measly $700 semester fee to study at a good university here in Switzerland.
Not a full ride of course, but a scholarship of some kind? Definitely. Lots of scholarship money is out there and if you have the ability to get accepted at Stanford you undoubtedly qualify for many scholarships.
Frankly if you are a student from a US high school accepted at Stanford and you don’t have a scholarship it’s because you probably made zero effort to get one.
Exactly what do you think you have refuted here? All you did was link to a financial aid page for a private university that costs nearly $90k a year to attend.
Every university has some sort of financial aid page or program. Every university has students helping to pay for their tuition and costs with scholarships, cash, loans, you name it…the university doesn’t care how they get their scratch, just that they get it somehow.
My daughter had nearly a dozen different scholarships and her degree still cost me some money…and she also chose to take out a student loan too.
This sort of thing keeps getting written. You will still be able to art, even if AI art becomes better. I can still, to this day, still go for a walk even though cars exist. Not because I need to get somewhere, necessarily, but for the love of walking.
This is only being litigated in this way I think because art people never thought automation would come their way, and they're thinking about this stuff for the first time. A silly comment from some GenAI exec is not much justification for such an article. People know that writing and making music is fun. Said exec is just paid to say the opposite.
>You will still be able to art, even if AI art becomes better.
I don't know, maybe all the artists, musicians, and poets will be too tired from doing whatever menial labor bullshit jobs are left post AI to create great art.
Why not direct our efforts into enabling more people to devote their lives to creating art instead of fewer? Wouldn't it be amazing if everyone was able to spend their time creating because the machines were doing the work we actually didn't want to do?
In the near future, we will have less art made by professionals and more art made by hobby artists (with or without AI support).
It will be a tough time for professional artists, but personally I don't believe we will "lose the human spark" in this transition.
If all AI art is soulless, new artists will rise that can put the missing part back in.
How many amazing potential creations have been lost because their creators did not have the time or money to make them? I'm looking forward to see what these people will create in the coming decade!
I think without financial support you're not going to get very good art. Some of the greatest artists of the renaissance were only able to spend all day painting because they had patrons. My city has lots of public art because the government spends money on it.
I'm struggling to understand this take because we already have the meme of the starving artist and their angelic patron.
What you're claiming is essentially that when tools make it faster and easier to create art, art gets worse.
In the past this has certainly not been true - every new tool has led to new and wonderful forms of art. I'm not sure why it would change now.
I believe there's a substantive difference between digital art tools like photoshop or illustrator and LLMs. LLMs do not create "art", they create images. Previously, you had to PAY an artist to create art. Now, you pay for an API to create an image loosely based on the specifications you give it. As a result, there is less art in the world.
Art models, to me, are very powerful paintbrushes. The technical skill of the artist is less important, but their ability to create art that makes people FEEL is more important than ever.
you said there was a "substantive difference" but you just gave an ungrounded definition?
I'm not quite sure how to define the difference, but I can tell you that LLMs do not create art, and never will.
They won't suddenly stop spending that money now that AI can help with the basics. Cultural expenses are a great way to buy reputation.
I think it's interesting how this always gets framed as reducing the number of creators.
I run an AI storytelling site with thousands of users and by most measures I wouldn't say it writes better than a practiced human, but what it does do is allow would-be consumers to become their own creators.
Some users spend $100+ dollars a month, not because it writes 10x better content than what a $10 Kindle unlimited subscription offers, but because even at 50% the quality they'd rather have the story they wanted to tell but didn't exist elsewhere.
For me AI art is not about replacing creators: it's about letting users go from passively consuming content made by people who have skills, time, and/or resources they lack to actively creating what they want to see in the world.
I'm not sure how compatible being against the democratization of creation really is with the narrative of defending the arts.
That’s it and with every invention the percentage stays the same 90% consume and 10% create. The insane amount that gets put into options of image models to create the maximum amount of images per hour is the dumbest approach possible for the technology at hand. Nobody is there that wants to consume it.
The reason why I'm personally convinced that generative AI for creative output will be bigger than any of the "PhD level agents" everyone is chasing is that gen AI can finally change "90% consume and 10% create", by changing what it means to consume.
People assume my storytelling tool is for writers because until relatively recently there was no other option: you couldn't make a writing tool for readers who only want to read, or a photography tool for viewers who only want to view, or a music tool for people who only want to listen.
But we're getting closer to tooling that makes it so frictionless to create, that you can insert the creation process in the consumption process, essentially turning it into curation.
-
Next week Gemini Flash comes out and we're going to have access to the first publicly available image model that can edit images with as much nuance and understanding as current SOTA LLMs do for text.
Everyone is focused on a possible plateau in intelligence, but steerability is the key for creative output, and there are so many angles to attack it from that we're at the infancy of what generative AI can do for creative output.
I recently got hit up out of nowhere by somoene who was like i'll pay you to storyboard a music video of a famous dead rapper. Think tupac hologram but somehow even worse in terms of digging someone up from the grave. I was like ok sure why not, I know this generative AI stuff. Got paid like a hundred bucks to start on this storyboard.
Thing is the story was so fucking stupid that I stopped after generating one image. They probably didn't make the story itself with AI but it sure was ridiculous, I basically quit as soon as they asked me to somehow include a bunch of jailed celebrities like diddy into the next storyboard image, getting out of it saying that it went against my legal capabilities to include real living people.
I hope you realize that some people with millions of dollars just have really head ass ideas they're going to put through your thing and things are going to get weirder sooner as a result. Like that idea would have ended at the drawing board with a bunch of people talking each other down from it not that long ago but now it's basically like that WKUK sketch about a grape out here [1]
[1] https://www.youtube.com/watch?v=mqgiEQXGetI
Side question, what percentage of your clientele has NSFW usecases?
> Some users spend $100+ dollars a month
IMO the real interesting questions your post brings up are:
- what kind of labor is the $100 your users are spending coming from
- where would these $100 have gone before your product existed
- where do these $100 go once they change hands from users to you
There are many answers, not a single answer, to these questions, but I think they're worth considering.
In a previous life I worked for a startup where, getting to know our users, I realized that if you broke it down the revenue was essentially money being funneled from hourly-wage workers to venture capitalists, which was a sobering realization.
Not one person in this world should have to motivate you to do the thing you love. This is a reckoning moment for a lot of people, artists included. Dig deep. Because the truth is that a lot of people do things, not for the love of and enjoyment of doing the thing in and of itself, but because of the satisfaction they get when they impress other people.
"You're so talented" "I love your art" "I wish I could do that"
And now there is that fear that the validation will go away. And what are you left with?
You and the thing you didnt really love.
Dig deep.
>Because the truth is that a lot of people do things, not for the love of and enjoyment of doing the thing in and of itself, but because of the satisfaction they get when they impress other people.
And there's nothing wrong with that. That's part of what makes art art. Creators are not solipsistically creating art for its own and their own sake: they do it for sharing it with other people, and getting the others' recognition and connection.
Some of the greatest rock music, for one, was written to impress the other sex - and some of the best dance music was written to see people dance and their bodies connect to it.
Machines already took almost all artist jobs a long time ago. Ever heard of the camera?
Machines already took almost all musician jobs a long time ago. Ever heard of recording?
The current winner-take-all art fields are not very healthy for humans to work in anyway, as there are infinity starving artists for every success. Eliminating the few remaining successes won’t really change much from the perspective of the average artist.
Almost all human art today is already done by hobbyists.
It's incredible to make the claim that almost all art is done by hobbyists considering how large the movie, tv and video game industries are by employee headcount.
Okay, compare that to the amount of hobbyists by headcount. Also subtract out the support roles in those industries that don’t constitute “artist” roles. Is a programmer an artist? Is someone working on color grading according to the specifications of someone else an artist? Subjective for sure, but at least for me most roles in those industries don’t qualify as “artists” since they don’t exercise substantial control over the art output.
As an illustrative comparison, if we apply the same “massive org chart” model used in film or video games to painting, each employee would apply a few brushstrokes according to requirements specified by a painting director. Is each of them an artist? Or is the painting director the artist?
Parent did say art.
Most of the big music, tv, and movies industry output is commercial crap, designed by committees and focus groups, and churned for the profits alone.
Meanwhile 95% of the music industry is either hobbyists or professionals that do it with a passion, and just barely make ends meet.
> Why not direct our efforts into enabling more people to devote their lives to creating art instead of fewer? Wouldn't it be amazing if everyone was able to spend their time creating because the machines were doing the work we actually didn't want to do?
People are also doing that. Note you're not travelling thousands of miles on foot to tell me this message; nor are you and your family doing subsistence farming. And you probably have time to do art if you want, because of existing advances.
You have missed the point of his article. These things keep getting written, by smart people, BECAUSE other smart people like you are not getting that point.
Obviously, people can still produce art. What he’s talking about is the corrosion of the fabric of culture that will serve to discourage and coopt the art-making impulse. It’s like this: When I was working on the last chapters of my book, I spent a few weeks in a cabin with no internet. Why? Certainly, the mere presence of internet wasn’t interfering with my keyboard, right? Right. But part of my brain wanted to surf and play games online, and it was dragging the rest of my brain with it.
“Art” is a cultural dynamic, not just a matter of production. It is being colonized by people who use the word and don’t care about art. And some people who might have become great artists will now become frustrated consumers and nothing more— their energies dissipated.
I don’t know how big this problem really is. But that’s his concern. At least engage with it instead of implying that this Stanford professor is a moron who doesn’t understand how computers work.
> I can still, to this day, still go for a walk even though cars exist.
In many places this is straight up false
Such as?
Orlando springs to mind, the pavement just siddenly ends and people who see a couple walking with a kid pull over and offer assistance, from personal experience.
>when we work at our favorite activities — cooking, gaming, hiking, music-making, writing (to name a few) — the process is often not easy or pleasant or smooth.
This gets to the heart of what bothers me about using LLM to emulate Art. This is the justification people make for using it, but no Artists I know finds making Art consistently unpleasant. Sure, there are times you're blocked or tired but generally in the end you are satisfied or have leaned something helpful from the experience. This is all of life. This is why there is Art. If you just want a pleasant time, you're looking for entertainment, not creative endeavor. Just put on a VR headset and leave making Art to the real Artists.
Art is like anything. If you want to get good at it you need to discover how to get dopamine from it and exploit that to habit and mastery. This is how you get in shape, it's how you improve your golf game, how you develop software, raise kids. Yet for some reason humanity wants a cheat code for all of this. Pills to make you skinny, video golf to give you an advantage at the links, top-tier schools to make your kids awesome. We're hellbent on dumping the baby out with the bathwater because we think the point is putting the tub away.
AI songs present as a refined end-product but you don't have to treat them as anything more than killing time or B-grade inspiration for something else. Instrument skills will surely be preserved somewhere due to nostalgia, it's the industry gatekeeping that is coming to an unharmonious end along with all the silly status games if AI songs ever take off.
For instance, why does an indy band need an AI-free backstory but a popstar can outsource all their imagination to the producer who may or may not use autotune or AI riffs?
Because the kind of people who listen to indie music are not thr same as those listening to pop music. Indie artists are valued for their originality or authenticity. Pop music is valued for the spectacle it puts on, not just the sound waves that come out of your earphones.
Personally I don't see any point in listening to AI generated music. It's not something I'll ever be interested in, certainly not something I'd be willing to spend money on
I have been moved by the rare piece of AI genre music. Admittedly, most of Gen AI that gets updooted is repulsive though. This emerging field is not for the faint hearted and has terrible curation chops.
"Instrument skills will surely be preserved..."
Citation needed.
are you kidding? AI will make live performance more valuable, not less. The human connection is what we want.
Citation: the magic feeling my instrument makes when it vibrates in my hands when I play. AI isn't going to take that away you soulless gremlin.
CITATION NEEDED, INDEED!
Are there any examples of popular artists playing instruments regularly at shows? Not formerly popular artists.
Nearly all the concerts I've been to recently were rap artists and DJs . Beyonce doesn't place a huge emphasis on instruments in her show, nor do the KPop artists. Do people go see Taylor Swift for the guitar? Sometimes someone will bring instruments on for a few songs.
Original composition still happens with instruments, usually but not always.
If you go to dj show and get mad that there is live instruments, that’s on you.
I am not kidding.
https://www.theatlantic.com/health/archive/2024/12/music-is-...
I have been a musician at this point for 35+ years. The quote from the Suno CEO is beyond absurd.
I have spent quite a number of hours with MusicLM in AI test kitchen. To me, that is actually a new form of music synthesis. Synthesis in the context of FM, subtractive, etc.
I suspect MusicLM had some tracks removed from the training data because it use to make absolutely wildly creative psytrance clips and now it just doesn't.
Suno on the other hand to me has basically been a complete joke. The training data is just not wide enough to do anything interesting and new.
I think the only way to really use these tools creatively would be to produce your own trained model with AudioLM. The creative use then is in the musical output and not selling the model as yet another SaaS for $x.99 a month.
Of course, you could use Suno to create loops and sample material. The way midjourney is the greatest thing possible for digital collage in Krita/Photoshop.
GenAI art though on its own be it midjourney, stable diffusion, Suno, MusicLM, whatever already jumped the shark for me a year ago. The problem ultimately is the output is just so very limited.
Technically, I feel that the current diffusion angle always has to fight against sound quality.
But maybe this angle can find its own "Retina" moment, and Suno's v4 is closer.
But the hard flaw (almost unsolvable) of this angle is editing.
I have been exploring methods based on DRL( https://github.com/chaosprint/RaveForce), but I think there is still a long way to go.
> Suno on the other hand to me has basically been a complete joke. The training data is just not wide enough to do anything interesting and new.
Hard disagree, Suno is terrific, but likely depends on what you define as "new". Of course AI models can't explore outside the data they are trained on, but they can effectively generate within the parameter space, including areas unexplored directly by the content it was trained on.
Have done some really interesting mashups of styles that people aren't doing and the results are great imho.
If you are trying to make it do very specific things like you would another instrument, it's not the right tool.
I would love to hear them. Would you share a link or two?
I really enjoyed the perspective of this author and the article's contemplative tone. I think an important perspective is that these are tools for consumers, not creators. While we are all obviously both to a degree, the current economic model makes a clear distinction.
When it comes to monetary exchange, we are either artists or consumers. AI art has the potential to create a massive imbalance between these two fields, reducing the friction of being a consumer, and greatly reducing the number of people who are able to support themselves as artists. If you view yourself chiefly as a consumer this might (very narrowly) be viewed as a positive.
The social cost however is creating a society that favors consumption over creation, and weights incentives accordingly
If you look back at the early days of broadcast television, for example, there were only 3 channels. 72h of programming at most in any 24h time period. This deeply favored the creator (or perhaps more accurately, the owner of the channel of distribution). Now we have YT and Netflix, both of which are biased towards consumption. Is it all bad? In some ways; we have to wade through more to find the good stuff. But in other ways, we have more to choose from and more people can be creators.
It's nice to see someone that gets it, and can explain why all of these generative-AI tools are completely pointless so well.
A gigging musician I know, who is involved in many projects, including leading an on again, off again BiG BanD, recently had another aquaintence, play a recording to him of the soul song that this person had "written" with ai help. I am struggling with comming to terms with the idea of artificial music. Some people like it, so it will stay around, and I cant begrudge them, at least for now, as ai will probably end up spawning new generas ,that would likely not come into bieng without it, which I suspect, the whole point will be inflicting there "art" on anyone they can. This artificial art, has no back story, no sweat, no triumph, and will never have a voice or style it's own, it's only and explisitly derivitive, but there are those who are in some strange way comforted by that, a bieger shade of ecru, free from angst and any possibility of it showing anybody up. Though all in all, I think it's, orrible, right nasty, and I imagine that it will end up banned like a lot of other artificial things, that prove to be bad, for everybody they touch.
edit:synthetic creativity
> AI music company Suno’s CEO was quoted as saying “I think the majority of people don’t enjoy the majority of the time they spend making music” in what sounds like yet another justification of Suno’s prompt-based (and very popular) generative AI music service...
> Such systems offer the promise of skipping the labor by bypassing the creative process and, yes, the difficulties, confusion, and frustration inherent in such endeavors, but…
The true purpose of AI is to free humans from creative and engaging work, so we can spend more time doing drudgery (like monitoring AI output).
> Statements like Suno’s perfectly captures the prevailing public mindset about AI: that Artificial Intelligence is little more than a labor-saving optimization tool. This mindset tends to be good for #Capitalism, but betrays not only a lack of understanding of why people make music, but also a profound lack of imagination regarding how we could, or would want to live with our technologies in our lives.
The true goal of Capitalism is the elimination of anything that isn't Capitalism, so it can be replaced with more Capitalism, until everything is Capitalism and nothing else remains.
Amusing to see a Stanford student complain about capitalism. Attending Stanford costs $70,000 a year. The bastardized John Adams quote: "I must study war so my sons can study business. My sons must study business so their sons can complain about capitalism."
Very amusing to read this comment which has nothing to do with the article. The writer is a professor, and the article provides a lot of insight.
I guess your father’s quote would be “I must read an article so my sons can skip it and make comments.”
Did you read the full article?
Of all the billions of dollars in the world wasted on disturbing excess, aren't you thankful that some of it goes towards something worth more than what money can buy?
Can you explain your logic here? Why couldn't the student complain about capitalism while being enrolled at Stanford? I would have rather assumed that the high semester fee may perhaps contribute to them having that opinion. I pay a measly $700 semester fee to study at a good university here in Switzerland.
"Looks don't matter" - good looking people
"Money doesn't matter" - rich people
The reason he's in a position to complain about capitalism is because he's rich. It's funny because it's predictable.
Large numbers of Stanford students are on financial aid and/or scholarships
I’d lay a bet down with good money behind it that every kid at Stanford with wealthy parents also has or had a scholarship or two.
Not sure that the presence of a scholarship, financial aid, or even a student loan necessarily proves poverty of the student’s parents.
It's your belief that everyone at Stanford gets a scholarship of some kind?
Not a full ride of course, but a scholarship of some kind? Definitely. Lots of scholarship money is out there and if you have the ability to get accepted at Stanford you undoubtedly qualify for many scholarships.
Frankly if you are a student from a US high school accepted at Stanford and you don’t have a scholarship it’s because you probably made zero effort to get one.
The data doesn't match your belief
https://financialaid.stanford.edu/undergrad/
Exactly what do you think you have refuted here? All you did was link to a financial aid page for a private university that costs nearly $90k a year to attend.
Every university has some sort of financial aid page or program. Every university has students helping to pay for their tuition and costs with scholarships, cash, loans, you name it…the university doesn’t care how they get their scratch, just that they get it somehow.
My daughter had nearly a dozen different scholarships and her degree still cost me some money…and she also chose to take out a student loan too.