Name: Brainstorm AI 2024: AI Music Finds Its Groove
Uploaded: 2024-12-10
Duration: 17 min 28 s
Description: Ed Newton-Rex, Chief Executive Officer, Fairly Trained Carol E. Reiley, Co-founder and CEO, DeepMusic.ai Moderator: Verne Kopytoff, Fortune

Ed Newton-Rex, Chief Executive Officer, Fairly Trained Carol E. Reiley, Co-founder and CEO, DeepMusic.ai Moderator: Verne Kopytoff, Fortune

Brainstorm AI 2024: AI Music Finds Its Groove 

Conferences

And a big thank you to our guests. We're here to talk about music. And before we ask, uh you know, what your organizations do, I wanted to set the stage and ask um uh A I can, can A I music be groundbreaking? Can it be as uh unique as the Beatles or Jimi Hendrix or Beethoven? It's a big start. I start with you Carol. Start up with a Yeah, an easy question. You know, I think um when I think about A I, it was um like through the full entire stack because I think the obvious one is like music generation and helping artists compose and write songs. Um Then I started like digging into like music production crowds and you looked at like the labels and then like the, the front of house marketing tools. And I, you know, I think the A I is, we agree is like the uh the superpower. So how do you use A I to become super creative? So for me, I was really looking at how do artists become super creative because I view them as like the most creative humans on earth who have made this their profession, how would they do it? Um And then see how that impacts um society and humanity later on. Um I think it has great potential. And I think for the last few years, I've been digging into exactly that question on where is the value and how can we make the pie bigger? Um And, you know, recent surveys have also showed that uh the number one job that young people want to do is to be an influencer. So how does that type of creativity also utilize A I to create content to put that out? What about you? What's your perspective? I, I think I've seen, I've seen innovative things be done with A I Music. Absolutely. When I was at Stability A I, our audio team released a product called Stable Audio and it was basically a sample generator and, you know, you could get it to do some pretty wacky things. Like one of my favorite things that one of our team produced was I think the prompt he used was hammering wood at 128 beats per minute. And like it came out with this weird combination of like a kind of hammering sound effect, like, but in like a drum beat form and I think sample generation in general is something where, where this tech is already being adopted and you can get some innovative stuff out of it. I do, you know, I think unfortunately though, like a lot of what is produced with a lot of the systems out there at the moment is incredibly derivative, right? People call it slop and I think that's like a pretty good term for it. So I think that the the majority is not going to be, you know, musically innovative in that way. I think uh well, I just also follow up and just also push that like A I could, the same thing can be said about humans. I think a lot of the stuff that we also produce at least to start with is also incredibly derivative or trying to copy cord. So I'm, I agree. I think that maybe A I could be used as a challenger to, to push us to be better. Um could be one lens to also look at. But I do feel like it's an inflection point where it could be a lot of sloth and, and derivative music and A I might just streamline everything to be the same derivative music. So, so I'm looking for us to really push that boundary. Well, tell us what your organization does. Uh Deep music. A I? Um So, uh you know, I think uh five years ago, II, I, uh I just sold my company which was in self driving cars which had a large impact on society and I was looking for what could be the next big disruption and I went through my own existential crisis. I'm like, what if a I reached the point where um it could replace human creature. These because we sat on panels and I always had said, if your job is in the creative, you're gonna be safe. And I was like, what happens if that assumption isn't true? Like, what would that look like? And what if we were to let a I replace the creative side, the stuff that I enjoy doing? What would that look like or how would we, how could I find a way to like, utilize and lean into a I so that um that could become a new area or like, so, so I really wanted to partner with artists and um people who were more creative than me. I'm an entrepreneur and I think that's like one set of creativity, but I couldn't figure out how to like systematically break down how I went through my creative process. I know that sometimes I just get inspired. The light went on my head, but I started talking to different types of creatives trying to figure out how they went through their process and had um landed with the San Francisco Symphony. I was working with them and composers and musicians trying to understand that. And so I created this nonprofit with one of my dear friends from grad school who is a artist, um Hilary Hahn, like it was during COVID, so she was not touring. And so we had like a three hour beautiful conversation about what happens if A I was to disrupt her career. Um because I had seen that in the self driving car space where eight graduate students from Stanford, we were able to put a car out in three months um on the streets driving around mountain view. So I would say I could see the acceleration of A I and I knew that it was coming for creative soon and wanted to help the artists not be reactionary. But if they were to lean into it and band together and understand the tools like and be part of the companies that were building these tools, how would, how would that change like the course of humanity? That was my 2018 revelation. And what I was been trying to unpack since then. Fantastic and tell us about Philly trying uh 30 well, 30 trains a nonprofit I set up about a year ago. Um I worked in, I've worked in Generative A I since 2010. Uh And for, for like a day for about 12 years uh since I was joined Generative A I to about 2022 people didn't train their models on copyrighted work without a license. Everyone knew it was illegal. No one did it. They, they trained research models but they would never commercialize it. But then in 2022 a few big companies, you've all heard of started releasing commercial models and threw caution to the wind and I ended up working at another generative A I company and, and eventually I quit that company because, because I, I thoroughly object to this unlicensed training. I think it's illegal and I think it's unethical. And so I set up fairly trained to, you know, we're a non profit. What we do is we certify generative A I companies that don't take that approach, we certify generative A I companies that don't claim that they can scrape the life's work of the world's creators and train vast commercial models on it. The good guys. Well, I think so, but everyone thinks they're the good guys, you know, and then you've also uh supported or uh uh had your petition signed by 12,000 creatives including this, uh the Cures, Robert Smith and radiohead singer Tom York, uh calling for uh compensation for artists. Yeah, it's tripled. We have 36,000. Um Yeah, this was a statement we put out like a month ago and it's a 29 word statement. The wording, it's still available for signatures. If you happen to agree with this, we're in San Francisco. So you probably don't. But if you happen to agree with this, the statement is the unlicensed use of creative works for generative A I training is a major unjust threat to the, to the livelihoods of the people behind those works and must not be permitted. That's the entire thing. It's been signed by a lot of, a lot of creators including you know, a lot of well known creators, but more importantly, like thousands upon thousands of working creative professionals who you probably haven't heard of and whose lives are being impacted by this uh this right now. So if you do agree, please go and sign it. It's still open, it's still rising. And now the music industry has sued a couple A I uh music engines. Um What should we think about that? Carol? Um IIII I agree that they need to be compensated. Acknowledged. I, I do think that uh technology has disrupted the music industry many times over from like sampling electronic music. Uh And this is another new tool that is coming and they need to be, I think, fairly compensated. Um I think it's great that they're standing up. Um And, and doing something about it. Um Another approach is II I think it's also interesting that some artists have openly model, like create models of their voice so that the public can utilize it and they can rev share things. So I, I feel like there's new business models that need to be created. We need to be experimenting with them, but I do 100% agree that the rev needs to be split. Um They, they need to feel fairly, not just monetarily compensated, but I think acknowledged for their work. Um There needs to be ways to like spot if that has been um not upheld. I think what Ed is doing is fantastic. Um And I, I think really banding the artist together to have like a community and a place to talk and have conferences like this is important. And what, what kind of uh compensation scheme do you envision? I? Is there a right one? I don't think there's a right one. Actually, I think there's like, I think often A I companies will turn around and say, well, come on then rights holders creators like, you know, what's the silver bullet for how this should work? And I don't think there has to be one like the beauty of licensing and licensing is clearly the solution I think here, right? The beauty of licensing is that like two parties can come together and figure out a deal and figure out what works for them. That's not going to be the same for everyone. Someone, one person is going to want an upfront payment, one person is going to want a revenue share stability A I when I, so I did licensing deals like this with creators and you know, and, and you know, we, we, we put in place a revenue share with, with music creators whose music we trained on. Um you know, and I think I think that kind of model can be great because no one really knows how much money a lot of these things are going to make, but a revenue share can work really well. Um There, there is this like there is this argument that A I companies often give in legal filings as well, you know, where they will say, well, you know, it, it's kind of, it's not feasible that we could pay everyone. These payments would be so small, like, kind of what's the point of paying people and to that, I say you've obviously not been a creator yourself and like, got small royalty checks from Spotify or whatever it is, right? Like, you know, like just because the payment is small doesn't mean a creator doesn't want it and it doesn't give you the right to take their work without permission. So that, that's my view. I think you've got to start from licensing and consent and you work from there and if people don't want to be involved because you're not offering them enough money, then you don't use their work. And it's as simple as that. Uh Carol, what will a I ultimately mean for the jobs of, of, of musicians, do you think? Got it? I do think, I think a lot about like the future of work and where people go and how we utilize our time. Um You know, it's, it's undeniable that it will be disrupted. There'd be a lot of jobs replaced, particularly um things related to maybe like music for commercials or royalty free music. Um That, that type of thing. I, I don't know how much joy those artists have created music, but there is a paycheck there. And, and I, I do think that, like, um those lesser creative, like, uh those type of jobs might be uh more easily replaced because it will be cheaper using like a I generated music or you could have personalized music in the own your own lyrics. Um But I do think there's always room for like true creators or like, I, I do think that more people will be jumping in. I feel like uh music seems, I, I've worked in many different applications and music seems to be really one that resonates with the majority of the public. I feel like music touches people. It is um gets them through hard times or they themselves want to express themselves a certain way. So I think there'll be more artists than ever before. Um And people would be able to use the tools and I, I think what we need to work on now is really the interface of these tools. Uh You know, if you guys have played around with a lot of the, the, the tools that are available, I think they're fun for a while, but the stickiness level doesn't seem to quite be there. And what I'm interested in is, you know, how do we get people to come back to the tools? Um You know, I think there is something with personalization. People love like A I bias for things that are personal for their life or their, their album um photo album and they want personalized music, but it doesn't seem to be something that has any stickiness right now. So I think a lot of work needs to be done on the interface side from and ed, what percent of the music industry do you think is doing it right now? Following copyright that you believe they should? Well, I mean, in the A I world there are lots of companies who are licensing all their training data. So fairly trained have certified 16 music A I music companies, companies involved in A I music who have, who don't use copyright work without a license. It's the, it's the absolute minority of companies in the music space who are training on the world's music without permission. And, you know, the two most prominent Asuna and Yio and they've both been sued for huge amounts of money by the record labels and I'm, I'm sure if other people follow their footsteps, they'll be sued as well. Um, so, yeah, a lot of A I music companies are doing the right thing. Like I encourage you to check them out. They're, you know, they're on our fairly trained website. Um, you know, II, I agree with Carol. I think that there are good use cases. Um, for some A I music, you know, I think, you know, one thing we saw recently, you know, was like Randy Travis, the country singer using, who had a stroke using, uh, using A I voice models to recreate his voice and be able to release new music. And there are these sort of empowering stories. But unfortunately, I think right now it's pretty unavoidable that the main use case of A I music will be essentially the commodification of music. You know, there isn't text to music models. I agree with Carol. I mean, that, that, that retention, that that kind of that real use case for a long time, people were thinking, where's the mid journey for music? Right? Loads of people use that and then a couple came along, you know, and how many songs about your trip to the coffee shop? Do you really want honestly? Like text and music is not, is not the answer. Um You know, so I think, I think there's a lot of work to do in the A I music world to find the really, the really good application. And unfortunately, as I say, I think commodification is gonna, gonna be the most scalable one and that's something I'm not, I'm not thrilled about personally. So, well, let's look ahead five years from now. Um Do you think uh we'll have an A I Taylor Swift? Is that possible Carol? Um I mean, I don't, I don't, I would say that the goal is never to like mimic the artist, right? I think Taylor Swift is a powerhouse and her job is which I get no store on her caliber on her love. Yeah, I, I think what is so amazing about her is that she is the entire full stack. It seems like from um songwriter to like CEO to like paying those bonuses and um to all the, the crew and everyone in running this company. Um So she's like an empire and I think that is interesting to see how she markets herself, how she connects with the community. Um And I think a I can help maybe an artist that uh doesn't have her efficiency, productivity, like really like use her as a bismol to learn all the different aspects of her job and maybe utilize A I if you don't have as powerful as a team um to be able to do some of that. Um So while that, that is interesting um to have we, as we've seen like a I avatars, um you can go to hologram concerts. Um You know, they like, I think Kim KK, Kim Kardashian for instance is interesting because she puts out so much content. So can you imagine like a superstar like Marilyn Monroe being um like just the the team behind her constantly putting out 24 7 content with her seat, see things and doing deep fakes. And now all the eyeballs of our world are watching like only five characters. So I see. So I see that future happening where there's only a few a few superstars and it becomes harder for independent artists to break into that level. And get the, to get the average consumer's attention. Um And I think what the music is working on uh besides the the machine of oops of the, the artist is what are new ways to utilize A I that you can't do by yourself. So we are looking at like neuroscience um um and brain activity, live performances and um and, and getting the mood of a crowd to affect the performance uh during the stage live composition, um visuals for, for music, robots, dancing with ballerinas, you know, I feel like new types of art forms and trying to figure out uh what is interesting that is in a gimmick that has fame power, that could be um a new way to appreciate art um as well as behind the scenes for like dynamic pricing and helping, helping um some of these like art houses thrive in this, this phase. Great. I want to thank our panelists for joining us and talking about the future of music. I'm looking forward to how it turns out. I hope the musicians get their due. Thank you very much.