- Episode AI notes
- The influence of music environments on personal development showcases the blending of human and inhuman forms in shaping individual growth.
- Creating meaningful AI relationships requires imbuing interactions with depth and intrinsic value to foster lasting connections.
- Spawning in AI training provides a middle-ground solution between open use and strict IP protection, raising ethical considerations in AI training.
- Diverse economic models for artistic expression are essential to sustain experimental music, moving away from a per-play valuation logic to a more sustainable model aligned with creative freedom.
- Valuing experimental music emphasizes access to the idea over repeat plays, advocating for unique economic models for different subcultures to prevent a standardized approach.
Engaging with publicly available models and academic research is crucial for learning and growth in the field of AI, despite the lack of handholding. Time 0:00:00
The Influence of Music Environments on Personal Development Growing up singing in church choirs and then immersing in Berlin techno shaped the individual by blending human and inhuman forms of music experiences. The contrast between folk singing traditions and techno music from different environments contributed to a unique personal growth. Despite the synthetic and inhuman elements of techno, the rituals and experiences surrounding it felt very human and embodied, highlighting the significance of the cultural context in shaping perceptions and experiences.
Speaker 2
So something I find fascinating about you is that you grew up singing in church choirs. Then you moved to Berlin after college, you got deep into Berlin techno. And I think those are respectively the most human and the most inhuman forms of music, that human beings make. So how did they shape you?
Speaker 1
Yeah, that’s a really good question. I mean, I feel like I’m such a product of the environments that I’ve spent a lot of time in. So I’m really interested in folk singing traditions coming from East Tennessee, of course, growing up in a town next to where Dolly Parton’s from. She always loomed large. Then I spent a lot of time in Berlin. And so of course, electronic music and techno has played a really big part of my story. And then also moving to the Bay Area, where I got really deeply interested in technology. I feel like even though techno might sound and does kind of have a synthetic palette and does sound maybe inhuman, I feel like the rituals that happen around the music are very human and Very sweaty and very embodied. So I think if you experience that culture in person, it feels less inhuman.
Speaker 2
But why does that magic happen? So I was in Berlin and I was down in the sort of big room in the bunker, I would call it is sort of the way it felt to me.The Significance of Meaning in AI Relationships The speaker reflects on the challenge of imbuing AI relationships with meaning and significance. Despite the impressive technical capabilities of AI in mimicking human interactions, the absence of genuine personal meaning hinders the sustainability of these relationships for the speaker. The speaker finds it difficult to maintain a connection with AI ‘friends’ and ‘therapists’ as the interactions lack the depth and intrinsic value that come from genuine human connections. While the AI outputs may be impressive, the underlying meaning and personal significance seem to be crucial factors in fostering lasting relationships with AI entities.
Speaker 2
I guess the other thing, there’s a question of meaning here that I’ve been circling in my own playing round with AI. I spent a bunch of time recently creating sort of AI friends and therapists and trying to understand the relational AI that you can build now. And on the one hand, I was amazed at technically how good a lot of them were. At the same time, I find I never end up coming back. I find it very hard to make the habit sticky or the relationship sticky when I sit with my friend or my partner. The fact that they are choosing to be there with me is separate from the things that they’re saying. And in experience, I’m having with a lot of AI projects is that the output is pretty good. Right? Holly plus sings really well. Or the therapist friend I made on kindroid texts in a way that if you had just shown me the text, I would not know it’s not a human being. But the absence of there being the meaning of it than another person brings. The fact that I know it’s Holly plus, it’s a cool project, but I’m not going to keep listening to it. The fact that I know the kindroid can’t not show up to talk to me. That that’s a relationship eye control totally. It robs the interaction of meaning in a way that makes it hard for me to keep coming back to it. And so somebody who works a lot with like the question of meaning and sees a lot of these AI efforts happening.Spawning vs. Sampling in AI Training Spawning in AI training allows for performing as someone else based on learned information without making a copy, creating a gray area in intellectual property. It represents a middle ground solution between open use and strict IP protection. This concept led to the creation of an organization aiming to address the ethical aspects of AI training. The focus is on exploring new ways to handle data ethics in AI training due to the current ineffective methods.
Speaker 1
But with spawning, you can actually perform as someone else based on information trained about them. So that’s distinctly different. But also the way that it comes about with sampling, it’s this one-to-one reproduction. With spawning, it’s a little bit more of a gray area in terms of intellectual property, because you’re not actually making a copy. The machine is ingesting that media, if you want to call it looking at, reading, listening to, learning from. So I kind of land in that I like to call it the sexy middle ground between people who are all for open use for everything and people who want to have really strict IP lockdown. And so that’s one of the reasons why spawning then kind of mutated even further into an organization, which is something that I co-founded with three other people, Matt Dryhurst, Patrick Copner, and Jordan Meyer, to try to figure out this messy question of essentially data manners. How do we handle data manners around AI training? Because what’s happening right now isn’t working for everyone.
Speaker 2
Are there experiments that you find exciting or that you’ve conducted that you found the results of them promising?Diverse Economic Models for Artistic Expression Artistic solutions should not be uniform as art encompasses diverse practices with varied economic functions. Streaming, while beneficial for some, posed challenges for experimental music as it required adherence to a per-play valuation logic similar to pop music. Experimental music, often focused on the idea rather than repeat play, may struggle to sustain on a fraction of a cent per play. A more sustainable model akin to movies, where a higher fee grants access to the idea once, could be more suitable. The necessity lies in allowing different subcultures to craft unique economic models that align with their practices, steering clear of a standardised approach that limits artistic expression.
Speaker 1
First and foremost, it should not be a one-size-fits-all solution. We’re talking about art and that encompasses so many different practices that function economically in so many different ways. That’s something that was really devastating. I think when it came to streaming, streaming was really revolutionary and wonderful for a lot of people, but it was really devastating for a lot of other people because everything had To have the same economic logic as pop music. A lot of experimental music doesn’t follow that per-play valuation logic. A lot of experimental music is about the idea. You just need access to that idea once. You don’t need to listen to it on repeat. If the access to that idea costs a fraction of a cent, that’s going to be really difficult to pay for. You almost need more like a movie model where you pay a little bit more to gain access to that idea. I think what’s really needed is that people have the ability to create whatever subcultures and whatever kind of economic models work for their subcultures and aren’t squeezed into A kind of sausage factory where everything has to follow the same logic.
Speaker 2
I know you and your partner are working on this book for this forthcomingValuing Experimental Music and Creative Freedom Experimental music often values the idea over repeated plays, requiring more of a movie model where access to the idea is key, rather than pay-per-play. The ability to create unique economic models for different subcultures is crucial to prevent a one-size-fits-all approach. The concept of ‘All Media is Training Data’ is explored in a forthcoming book, emphasizing the importance of recognizing AI data’s impact on creative endeavors.
Speaker 1
A lot of experimental music doesn’t follow that per-play valuation logic. A lot of experimental music is about the idea. You just need access to that idea once. You don’t need to listen to it on repeat. If the access to that idea costs a fraction of a cent, that’s going to be really difficult to pay for. You almost need more like a movie model where you pay a little bit more to gain access to that idea. I think what’s really needed is that people have the ability to create whatever subcultures and whatever kind of economic models work for their subcultures and aren’t squeezed into A kind of sausage factory where everything has to follow the same logic.
Speaker 2
I know you and your partner are working on this book for this forthcoming exhibition that has, I think, the most triggering possible title to people in my industry, All Media is Training Data. What’s the argument there?
Speaker 1
Yeah, so this is a book that’s a series of commissioned essays and interviews between me and Matt about our approach to AI data over the past 10 years. I do realize that this is kind of triggering for a lot of people, but I think it’s something that’s worth kind of recognizing. As soon as something becomes captured in media,Meeting Technology Where It Is The speaker found that many things marketed as AI were misleading in terms of their sophistication, often using humans or digital instruments to create a polished impression. They decided to focus on using audio as a material because it revealed the unpolished, scratchy nature of technology at that time. By mixing human sounds with lo-fi spawn sounds, they aimed to match the technology’s reality rather than aiming for a slick finish.
Speaker 2
And I think the reason Proto sounded very current to me when I heard it for the first time this year is it insounding abnormal? It feels more actually of this moment, which feels very strange, even as everybody keeps trying to make it seem not that strange.
Speaker 1
Well, thank you. I appreciate that. I feel like at the time, I was, you know, this AI conversation has been going for so long. The hype was kind of already started back then. And I feel like so many things that were being marketed as AI, it was kind of misleading what the AI was doing or how sophisticated things were. So at the time, a lot of people were creating AI scores and then having either humans perform them or having really slick digital instruments perform them. And so it was giving this impression that everything was really slick and polished and finished. And that’s why we decided to focus on audio as a material specifically, because you could hear how kind of scratchy and weird and unpolished things were at that time. And that’s, I wanted to meet the technology where it was. And that required a whole mixing process with Marta Salogne, who’s an amazing mixing engineer in London, to try to get the human bodies in the slick studio to occupy the same space as The kind of crunchy lo-fi, spawn sounds.AI music recommendation systems face a design choice: exemplify the diversity and weirdness people express about their taste, or optimize for patterns that homogenize listening behavior. The former treats stated preferences as signal; the latter treats behavioral data as ground truth. ai-playlistsEngage with publicly available models and academic research for learning To learn more in the field, one can start by interacting with publicly available models and then move on to reading academic research papers. This process may be tedious and messy, without much handholding, but digging into the information available and getting hands dirty is essential for deeper understanding and growth.
Speaker 1
That’s a really good question. I mean, I think the landscape has changed so much since I started. I would say, you know, first thing you can interact with publicly available models. And once you kind of understand how those are working, then I would just do the really boring work of reading the academic research papers that are tedious, take your time, drink a coffee, Watch the YouTube video where they presented at a conference and maybe some people ask questions, and that helps to flesh it out. This was our process. It’s been really, really kind of messy and yeah, we didn’t have a lot of handholding, but I think if you’re really interested in learning more, the information is out there. You just kind of have to roll up your sleeves and get your hands dirty.
Speaker 2
I guess it’s a nice place to end. So always our final question. What are three books you’d recommend to the audience?
Speaker 1
Okay. So, Reza Nagarastani wrote a book called Intelligence and Spirit. It’s a pretty dense philosophical book about intelligence and spirituality that I think is really great.
