Written by Katie Chiou

IN CONVERSATION WITH is a series from Archetype where we interview artists in/at the edges of crypto across music, visual art, design, curation, and more.

Daniel Keller is a “former artist,” internet theorist, writer, and entrepreneur who has exhibited internationally as a solo artist and as half of Aids-3D, a seminal post-internet collective. He was formerly co-founder of New Models and Channel.xyz. In 2023-2024 he collaborated with and led "cultural strategy" at Vaporware, an Urbit-ecosystem startup.

He is currently building Cloudy Heart with artist Jon Rafman and leading Metropolis DAO, an AI investment DAO that is part of the Tribute Labs network.

The following interview has been edited and condensed for length and clarity.

Katie Chiou: For those who may not be familiar with your work, can you share more about your background/journey as a designer and artist? 

Daniel Keller: I went to art school, and then I was an exchange student in Berlin where I ended up staying for 15 years. I didn't actually graduate art school. My career took off pretty early on when I was very young. Frankly, I was too young to, I think, truly take advantage of the attention that I got.

I was working collaboratively with Nik Kosmas on a project called Aids-3D. We were part of the first wave of net art that turned into what became called post-internet art. One of our pieces was called OMG Obelisk which got curated into the first New Museum Triennial in 2009. Our career picked up very rapidly from there. After a few years, Nik and I stopped working together and I had a solo art practice for a few years after. The themes were very generally about technology and its effects on society, interpreting techno-capitalism, California ideology, startup culture, internet culture, etc. I would say a lot of my art came from a place of technoskepticism, which was very typical of Berlin culture.

Over time, though, I grew more optimistic about technology. I was increasingly frustrated that within the art world, especially within Europe and the particular scene I was in, there was only one acceptable stance towards technology—reflexively critical, negative, antagonistic. And that voice became adopted by the media in general. I saw early on that technology was becoming more of a partisan identity. The art world itself also was changing a lot, and people were less interested in my perspective in general.

I stopped making physical art around 2016, but I launched New Models with Carly Busta who used to be an editor at both Artforum and Texte zur Kunst, and her partner, Lil Internet. New Models formed out of not being able to have the open discourse that we, as artists and writers, wanted to have in traditional media forums. Other leading figures of that general scene that come to mind are Brad Troemel and Joshua Citarella, who transitioned from art into content and discourse fluidly and made it into his practice. I think of New Models as a part of that turn. My process of art went from originally interpreting these conversations into quasi-sellable objects into just publishing that discourse directly.

From there, a bunch of figures from that scene and I collectively started Channel, which was an attempt at building a composable media network using token-gated RSS as the primitive. The company structure ultimately fell apart but we open-sourced the platform and the site is still functional and churning out new content for token holders.

A little while after that, I made an anonymous NFT collection that was a lightbulb moment for me. “Okay, this actually is still a really vital medium for doing permissionless creative culture and having an instant market for ideas.” That somewhat circuitously led me into the Urbit-adjacent ecosystem, which I became immersed in during 2023-24, while doing ‘cultural strategy’ for a now defunct company called Vaporware.

KC: Can you speak more to the moments that made you more optimistic towards technology? How has your thinking about the intersection of art, politics, and technology continued to evolve?

DK: I think I was very much like most people in the 2010s that believed in a sort of neoliberal “end-of-history” kind of mindset about things. I felt like the path was laid out, and it was a very narrow neoliberal future. Because I was in Berlin at that time, I was adjacent to a lot of the early crypto and Ethereum scene. But I was really, really skeptical about the need for decentralized alternatives to institutions. Bitcoin and illicit commerce seemed like the killer app, and the only killer app that made sense to me.

After the US election in 2016, I was shocked, as were most people. I began to believe in a more higher-variance future. That’s when I started buying Ethereum and Bitcoin as investments. ICOs was another thing that really made sense to me as another killer app for Ethereum. Seeing the massive performance in crypto assets made me more generally optimistic about the viability of more-or-less making a living from being extremely online. Of course, then round-tripping all that money that cycle was a huge lesson as well. Just the understanding that money can flow to you, and things can happen, and capital can both form and dissipate was really eye-opening to me.

The last one I’ll mention is when I started using GPT-4. I was not technical at all, but I could do basic coding tasks for the first time. Scripts and stuff like that, super basic. That was when I did that anonymous NFT collection. That’s when I realized that many of the visions and promises of technology I originally criticized in my art practice were actually on the table now. That has really become a dominant force in my life, this sort of unconditional accelerationism and belief that techno-capitalism is accelerating towards some kind of singularity. We can shape it somewhat, but for the most part it’s driven by forces beyond our control. I am trying to make sure that whatever I’m working on now is aligned with that general belief.

I do think we are in the early days of the pre-singularity or very close to a singularity. That means a lot of different things to different people. But I do generally think that it’s the single most important and interesting thing happening in the world right now. I also believe that crypto has always meant to be the substrate for AI agents to interact with each other and with the general economy. I am interested in helping make that happen.

KC: When did your interest in both crypto and AI and its intersection really click for you?

DK: I used to be very skeptical, early models of Midjourney were sorta aesthetically interesting, but it wasn’t clear to me that they’d advance like they have. Seeing GPT-4 and Midjourney V5 in early 2023 made it undeniable that we were deep into some sort of exponential curve. I also need to give a shoutout to Mat Dryhurst and Holly Herndon here, who have been at the forefront of understanding how these types of technologies are going to be intertwined from very, very early on, before that was apparent to anybody else in my circles. I give them a lot of credit for inspiring my confidence in pursuing this general orientation, despite both terms being extremely divisive within the culture sector, to say the least. But, in terms of how crypto and AI are interacting right now, I think it’s sort of vastly untapped still. There are some exceptions that have been interesting style-wise, but for the most part there’s an overly agreeable nature of LLMs that does not make very compelling content. But, eventually, I think there will be orders of magnitude more AI agents using crypto than there ever were humans. I think the intersection of crypto and AI basically answers the sort of eternally unsolved consumer crypto paradox of what is the use case and where are the users?

It’s almost like humans were pen-testing crypto. We were beta-testing various primitives and stuff, but we’re not the end users. We won’t be the users. And to me, that gives me some comfort, I think, actually. 

KC: That’s a good segue into asking who is Cloudy Heart? What is your relationship to her?

DK: Cloudy Heart is an art project by Jon Rafman and other collaborators. Cloudy Heart is an AI-generated pop star, streamer, e-girl. I was brought on to help with the crypto-facing strategy for the project. It's primarily an art project, but also has a memecoin which we fair launched on pump.fun, and which within Cloudy lore, was actually launched by herself, to enable her progressively increasing autonomy. The token serves as schelling point for collaboration and storytelling. In clinical terms, I think of Cloudy Heart as a sort of attempt at connecting these two flywheel ecosystems, music and memecoins, that are very well-trodden and optimized. Within music, there is a sort of industry-plant/payola system—where Spotify playlists and astroturfed TikTok virality—determine success. Obviously, a lot of that is organic and simply harnesses organic virality and it's not so exactly top-down, but there's a system there. And of course, there’s a system within crypto of KOLs, market makers, etc. I think of Cloudy Heart as many things, but at its economic core, an experiment in connecting those two flywheels together and seeing what can emerge from that conduit. That is my perspective about things in general. Part of my former-artist mindset is that I'm really interested in the systems engineering layer of cultural phenomena—maybe more than the aesthetic or narrative layers.

But also because Cloudy Heart is an AI, she's self-improving. The technology that is forming her is constantly improving. She was not very autonomous at all to begin with, but is becoming increasingly autonomous. The sort of North Star for the project is that as she kind of gains more features and autonomy, eventually she's set free and she is an actual autonomous agent.

There's going to be a lot of legal work around that that I don't think there's very much precedent for. That's a whole aspect of the project that I'm really interested in. I don't think there's any case or precedent for agent culpability and agent personhood. I think that there's a good chance that agent personhood will happen, because corporate personhood already exists, and it's a short conceptual leap to extend those rights to autonomous corporations.

KC: Do you think there’s a clear goal for Cloudy Heart?

DK: Jon has a lot of plans for extending the narrative universe of Cloudy Heart and other characters that he is developing. I think of Cloudy Heart as a form of experimental transmedia entertainment. I think of it as part of a new class of tokens that are sort of post memecoin, I call them “attention coins.” I think this is a new formula for tokens in general that we are going to see a lot more of, a content stream attached to a token as opposed to a single image/meme. AI content is not inherent to that formula, but it obviously makes it a lot easier to generate cool content if you know how to use the tools effectively.

In terms of agent utility, she doesn’t have much yet, but we definitely have interesting experiments we want to pursue—have her collect NFTs, have a wallet, create generative music. 

There’s this Mat Dryhurst and Holly Herndon quote, “all media is training data.” Everything we’re producing can go into a huge proprietary dataset that will allow Cloudy Heart to become more autonomous over time. I think that is what is going to define the effective use of AI versus slop. The ability to curate and assemble proprietary datasets and actually apply taste. "Taste" is a buzzword, but I do think that actually is a real thing—that that matters, and will matter more and more.

KC: What about a target audience?

DK: I don't think it's like a focus group type of thing. I think there is an eternally compelling thing about a female face—that is, obviously, you know, connected to the male gaze—but is more fundamental than that. I think facial recognition in general is something that's just so lizard brain. A lot of the other AI agents didn't try to do a photorealistic personhood, which seems kind of a no-brainer enhancement to me. Most of them are anonymous, faceless dudes. I don't think they managed to penetrate culture because they don’t have that sort of tangibility. You need that sort of meme of GPT that’s like Shoggoth the monster with the little smiley face RLHF mask.

There’s obviously a long lineage of projects that are very similar to this. Lil Miquela is a very obvious one, Hatsune Miku. There was Ann Lee from Philippe Parreno and Pierre Huyghe around 2000. So I think it's made very much with cognizance of all of that, and attempting to move it forward—basically with the acknowledgement that a lot of cultural change is downstream from technological change. And even compared to the Lil Miquela era, the ability to do things that move beyond the struggle of the uncanny valley—both in terms of music and video—are there or very close to there. That unlocks a whole bunch of new narrative possibilities.

KC: Do you think the Cloudy Heart team is particularly preoccupied with token price?

DK: I think launching a token with a creative project that has long-term aims is such a double-edged sword. I think everyone is pretty aware of that. Having a real-time attention market indicator of success is hard if you’re not used to the volatility of crypto. It’s hard to get used to that without going through a few cycles. That being said, I don't think there's any other way we could have just launched something and immediately had the attention and a market and the ability to bootstrap initiatives.

The reality, though, is that if you’re trying to do something in entertainment, entertainment world timelines and memecoin timelines are just not the same. A lot of that comes down to informing the community and reassuring them that things are happening behind the scenes, even if there aren’t constant market catalysts.

KC: Switching gears a bit, you’ve spoken a lot in the past about the purpose and power of memes in propagating certain information and ideals. A lot has happened technologically speaking since those original talks in circa 2017, how do you think the purpose of memes has evolved?

DK: At the time of those talks, I was trying to channel the memetic power that was largely leveraged by the right for the aims of the left. This eventually sort of turned into “altwoke” and “dirtbag left.”

I definitely believe in some form of meme magic. I do think hyperstition seems to keep happening.

We haven’t quite seen real-time attention markets for memes fully play out. But I do think that the understanding of memes as a form of currency, and the fact that there will literally be these direct paths of monetizing them in various ways—that feels to me like a glimpse into what the new cultural form is going to be. I think seeing the rise of app-based sports gambling and prediction markets shows you that there's going to be a speculative element to the way that people consume culture that’s going to become more and more baked in.

I think of Cloudy Heart as very much being in this new cultural form milieu. The fusion of the attention economy, AI, gambling—it's not quite related to what you're saying in terms of meme magic, but I do think that's how memes are evolving into an internet-native culture form.

I haven't actually plugged my theory of “sloptimism,” but that’s what I think this is about. We're going to have an acceleration of slop. Within that deluge of slop is going to be the seeds of truly avant-garde, genuinely great culture. I believe that. We haven't quite seen it yet, but I do think it's going to be the new internet-native cultural form of post-art. There might be a crypto component, there's going to be a generative component, there's going to be an interactive, relational, aesthetics, network spirituality component.

KC: I don’t think the idea that there is genuinely good work in slop is a hot take, but I think where people diverge is how the future here is distributed. Who will have access to, be able to sort through, the slop?

DK: You will require some kind of cognitive defense mechanisms to not be turned into pure lizard brain content consumers. Obviously, that’s already happening. It’s more so an attitude you can take where you have a sort of optimistic fatalism, find the bits that are genuinely funny or entertaining or whatever else.

Democratization is a risky word but, as an example, I think the barrier to make prestige TV is something I’m really interested in. There’s an indie film industry where you can make a film for $1M, but there’s not an equivalent for prestige TV because the startup costs are just 100x as much, and then you have to worry about renewal and all these things. 

I think industries like that are really stuck in local minima because of the cost of production, but if there are sloptimists who know how to use powerful tools, that could unlock whole new types of culture as well. That being said, I think a deluge of AI content will likely end up devaluing every medium it touches. And already, I don't think kids care about films or TV very much compared to millennials or boomers. But I think the only way out is through, and I am truly optimistic that the net result will be worth the disruptions and devaluations and that AI is going to help us create extremely cool and novel cultural forms. And as an unconditional accelerationist, I don’t think we have much of a choice.

Disclaimer:

This post is for general information purposes only. It does not constitute investment advice or a recommendation or solicitation to buy or sell any investment and should not be used in the evaluation of the merits of making any investment decision. It should not be relied upon for accounting, legal or tax advice or investment recommendations. You should consult your own advisers as to legal, business, tax, and other related matters concerning any investment or legal matters. Certain information contained in here has been obtained from third-party sources, including from portfolio companies of funds managed by Archetype. This post reflects the current opinions of the authors and is not made on behalf of Archetype or its affiliates and does not necessarily reflect the opinions of Archetype, its affiliates or individuals associated with Archetype. The opinions reflected herein are subject to change without being updated.