An Interview with Dmitri Cherniak

Jason Bailey: I don’t think there were many generative artists exploring crypto before you, right?

‍Dmitri Cherniak: When I saw Autoglyphs (2019), I thought: “that’s how I should be doing this.” Except, I wanted to use JavaScript tools. At the time, I was rendering individual frames or outputs from the code running on my computer. I would write an algorithmic system, but instead of capturing it in its full diversity, I might just mint one or two of my favorite outputs. It felt like something was missing because the edge cases and nuance of all the possibilities weren’t always visible to observers. 

Back then, I was minting on SuperRare, but I felt constrained by the fact that I could only upload image and video files, which limited the possibilities for interactive and live-coded works on the platform. To highlight generative art and help people understand it in context, we needed more. I basically asked SuperRare to make Art Blocks but they said they weren’t interested. So I started working on an idea where I would take a Raspberry Pi, put the code on it and lock it down entirely. But I knew we’d need some provenance to understand where the works originated — that was where the blockchain fit in. 

You would press a button, an artwork would generate, register on the blockchain, and after a certain number of presses it would destroy itself. 

In 2019, I met Sofia Garcia from ARTXCODE. We messaged on Instagram and decided we wanted to help make generative art a “bigger thing.” We were both so surprised at how little people understood about generative art, and we thought it could be well received by a wider audience. We started putting works into shows like CADAF in New York and Miami. Things were going well — my works were selling and people were learning about generative art. There were a lot of really great artists making incredible work, including Helena Sarin and Manolo [Gamboa Naon]. We felt like we were on a roll, and then COVID happened. 

That prompted me to reflect on what exactly an NFT was. Was it a collectible to go alongside my works or was it the work itself? Ultimately, I determined that the NFT was the most native form of the work, so that was what I considered to be the work of art. 

The way NFTs were presented at Christie’s in the form of Robbie Barrat’s AI-Generated Nude Portraits — the so-called “Lost Robbies” — also showed that NFTs definitely worked. But, for me, the most natural rendition of the piece was a tokenized form of the output created by the code. 

Things heated up over lockdown and MOCA bought one of my works for around 10 ETH, which at the time was around $4,000. Led by Colborn Bell and Pablo Rodriguez-Fraile, at the time MOCA were taking the world by storm, bidding on almost anything on the low end, but also bidding higher for works by artists entering the space whom they regarded seriously. They did research on each artist and even called them up to learn about their work. That energized many to engage with crypto, while also showing more established artists who might command higher prices that there were collectors here who took them seriously. 

However, at that time, code-based art still wasn’t well-known in the crypto community other than by the Autoglyph connoisseurs. The first time I came across Art Blocks, crypto art was starting to build some momentum outside of its small but passionate niche. Snowfro reached out to me on Twitter and introduced me to the Art Blocks website, which was still rudimentary at the time. 

My reaction was: “I’m going to show you a project that I think is going to help people learn about generative art. Then all the people that I’ve been making work with online for the past five years like Manolo, Matt Deslauriers, and Tyler Hobbs will see it, and then you’ll see another level of generative art.” 

And it worked. I think that Ringers was the right project to light the fuse for a lot of people because of how different the outputs could be. It’s also a very simple system, one that I spent three whole years playing with. I’ve made works using that system that are much more complicated and which, arguably, stand out more aesthetically. But at that time, on Art Blocks, I didn’t know how many people were going to collect them. The idea was that you could mint more than one — maybe you would get a fancy one, a regular one, maybe one with a color flourish or one that’s a little dull — and then put them all together and curate your own grid. 

After it launched, we shared the grid on OpenSea. I think Ix Shells shared it first, which helped a lot of people to learn about the project, and I spent a lot of time curating and highlighting different sets of Ringers. The community also began making unbelievable curations, including a collector called keight. I think the fact that the project sat in between punks and art lit a fire under the generative art movement. Ringers minted out in 18 minutes, and there were 1,000 of them. At the time, I decided not to mint any Ringers myself even though I really wanted to, because I wanted people to have the chance to engage with generative art. 

For me, the original Art Blocks minting process was the ideal way to mint generative art because you could do it in a leisurely way and see a number of outputs. But then it basically became a slot machine where people would mint as many as they could in order to get a rare one — then when they did, that would cover the expense of minting all the regular ones. But that also made it gambling rather than a process where you could mint individual outputs and appreciate every single one. 

‍JB: I think what was important for crypto art collectors in understanding generative art was the quantifiability of traits. 

‍DC: With Ringers I thought about traits a fair bit. I even minted all 1,000 pieces on a testnet in order to see what the distribution of all the traits would look like as a complete set. 

I regard the entire set of Ringers as a much more compelling piece of artwork than any specific Ringer. The distribution of traits across the set is therefore very important.

‍JB: Abstract and generative art are intimidating to a lot of people with no background in art. Was there any element of your background that made you inclined toward generative art?

‍DC: My background was in engineering and I was good at math, so I was basically told I had to do that. But, as a kid, I also loved to make art and I always appreciated the creativity that went into math problems. I regard automation as my artistic medium because before I was making visual art I was solving weird problems with automation. My role was somewhere between an analyst and an engineer, but that gave me “value” in this capitalist system. There would be a problem, and I would come in and spend a bit of time researching it, and make the problem go away by automating the system.

‍JB: What was your early art like? Were you making analog art, computer art, or a bit of both? 

‍DC: When I was a kid, I did a lot of detailed pencil drawings. It was highly detailed sci-fi stuff in the style of Beast Wars: Transformers, that kind of thing. Even after I started sharing my work online in recent years, my dad would tell me it reminded him of the work I used to do as a kid in elementary school. 

My first code-based art was Taylor Swift ASCII Art (2009-2015). Back in 2009, I had to build an AI game for a course at university — a variant of Go before AlphaGo beat all the humans. I named my game “Taylor Swift” and my friend named theirs “Miley Cyrus,” and it was a contest to be the most dominant AI. After my gamers got beat by Taylor Swift it would show her blowing a kiss in ASCII art, relying on a pretty simple form of image transformation. For the term paper, I made a stylized version in grayscale for the cover page. 

Then, in 2015, I rediscovered my term paper and posted the cover on Facebook. 20 people asked if they could buy a poster of it. 

I’d made enough things online to know that when people give a shit about what you’re doing, then you’re probably onto something. So I just started making more and more Taylor Swift ASCII Art in my spare time, refining the toolset and doing some video transforms. I was living in Oakland when the curator and artist, Patti Maciesz, asked me whether I knew about the history of Andy Warhol, Sol Lewitt, and other artists who had automated outputs in the past. I ended up doing a big solo show of Taylor Swift ASCII Art in 2015, which actually introduced me to crypto because my work was bought by the founder of a big crypto exchange. I explained my process to a number of tech entrepreneurs and they got it. I think they appreciated how I could use engineering, which is not generally regarded as a form of creativity, to generate art. 

At the time, people were telling me that my work resonated with them on an emotional level, but it was also generated in a fraction of a second, and if I didn’t save the random seed, I’d never see the output again. My Taylor Swift ASCII Art remains one of the projects I’m really proud of, not least because it followed a multi-year arc that made it highly cohesive. I feel the same about Ringers and Light Years (2022).

‍JB: Your patient, multi-year approach to Ringers feels opposed to the way most people have chosen to operate in the last two or three years. Many artists have been speeding things up in order to take advantage of a fleeting moment, but you’ve done the opposite.  ‍

DC: After Ringers launched, it was insanity — like a year’s salary — so I didn’t have to think about selling art. I soon started getting offers that honestly seemed great, but I was much more interested in keeping my practice cohesive by making long-term art projects than in selling. Ironically, that was good for my sales — I think people felt some confidence in that approach. One long-term goal of mine has been to have more generative art in museums. Despite the fact that the population is becoming more technologically savvy, until recently generative art was still not considered part of the creative realm. 

‍JB: I’d like to fill in the gap between the “Taylor Swift” work back in 2015 and the moment you minted your first NFT in 2019. What were you working on over that period?

‍DC: I was developing my toolset, working with geometric forms. I also got into plotters. I don’t use Processing but instead built my own thing based on top of React. All of my stuff is generated in the browser and it’s mostly vector graphics because a lot of the time I’ve been focused on trying to make scalable prints. I was pretty lazy, wanting to be able to scale my work up and down without having to worry about pixel resolution. A lot of my work in that period was inspired by Manolo, as well as Jared Tarbell and Zach Lieberman. The speed at which they were iterating was remarkable. Jared Tarbell’s Substrate (2003) is one of the greatest works of generative art — it’s actually my favorite generative work — while to this day Zach Lieberman is the most influential generative artist for me. A lot of forms that I’ve played with, he played with first. 

Once Trump got elected, I literally didn’t make art for a year.

Then, in 2019, I made it my New Year’s resolution to start sharing more work online. So I started using hashtags. That’s when I connected with Matt, Tyler, and Manolo, and started putting my work back out there. In that period, I was also learning more about Constructivism and rule-based art. I wasn’t a graphics coder but what you would call a full-stack developer — backend, frontend, deployment security. I’ve also been the CTO of a startup for the past 10 years. 

‍JB: Probably one of my favorite works of yours is your collaboration with Helena Sarin, gen2GAN (2020). At the time when that project came out, I think GAN art was starting to get a little repetitive. I think we’re at the same place with prompt-based AI art today. But I’m interested in how artists find ways to break that monotony or repetition. I thought it was a gorgeous solution to explore GAN art and traditional generative art together. 

‍DC: I completely agree. I think that right now, with the recent influx of prompt-based art, where you provide a sentence and an AI generates the imagery, you’re seeing a backlash where people are like, “AI art isn’t real art.” I actually have a book due for release soon that is the counterpoint to that — the clearest artistic exploration of using AI. I’m excited to get it out there. We’re doing a small run of only 200 copies. One of the cool things for artists about this crypto thing is that you can do it all yourself. 

Helena and I followed each other online and we both shared this aesthetic appreciation of Russian art. We were talking about it at the beginning of lockdown, and then I pitched her the idea because she was my favorite artist using AI — she was doing everything on her own scale with her own source material. She was owning the entire stack. The outputs weren’t these hyperrealistic things, they were very much her own distinctive style. It’s still such an honor for me to have collaborated with her on that project. 

People in the crypto space don’t talk about it because we only made three NFTs. But it’s not about the NFTs, it’s much more about the process.

‍JB: I’d also love to know more about your recent collaboration with the estate of László Moholy-Nagy, which freshened up the conversation around generative art last year. It also tied computational practices back to modernism. Tell us about that experience.

‍DC: Someone reached out to me originally, and said: “do you want to do a collaboration with Moholy-Nagy?” And I was like: “I can’t do a collaboration with Moholy-Nagy now that he’s dead — that doesn’t make any sense.” But I love Moholy-Nagy. I’ve read a number of his books; I have his manifesto. The thing that always resonated with me was his interest in using technological tools to promote art rather than for political or economic gain. So I said that I’d love to talk to the estate because, at the time, a lot of people were still looking to understand NFTs and how they work. 

It was really amazing for me to meet his daughter, Hattula. We think that he probably would have done this project if he were around because it was kind of his MO to play with new technologies. With this project, I wanted to connect art history of the 20th century with generative art today — to show that even though we’re generations apart, it’s not actually so far away. That’s how the name, Light Years, came about. 

I worked with Fellowship for the project, and we ended up exhibiting the works as silver gelatin renditions at Paris Photo as a way of showing the process. It required a six-month investigation before I had the final algorithm. But I felt really happy with the ultimate outputs because, in the same way that Moholy-Nagy took photography outside of the camera, I wanted them to register as code-based. The tour groups that passed by the booth at Paris Photo were able to engage with these works as examples of both generative art and NFTs.

‍JB: The different variations on Ringers — the Dead Ringers as well as The Rinkeby Sessions — forced collectors to decide which part of the creative process they really valued. You also did The Eternal Pump (2021), which followed soon after Ringers. Is there an element of social engineering here?

‍DC: There are two things — one is the actual execution of the code to create the output in the Art Blocks sense. That is what I really resonate with. But then also there are artists using technology to empower themselves. The idea behind The Rinkeby Sessions was that you could now get this art for your home, you could learn about Ringers, and you could decide where the value lay in the project. Is it in the beauty? Is it in owning a digital good? 

When I think about the social dynamics surrounding these projects, you have to understand that certain elements are a response to people treating me like an object or a way to make money. One of the best financial decisions someone has ever made in their life is minting a bunch of Ringers. That is crazy to me. Why do I have that responsibility? I’m very fortunate, but it’s also hard. As a human, I use my art to push back on the idea that I am just an instrument.

I don’t know how I’ll ever be able to top the Dead Ringers. It was insanity at the time. I took a look out into the world: we were in the middle of winter; it was COVID; the food banks in New York were in serious trouble due to food scarcity. And then all the while people were focusing on this super rare digital object. I used the energy that I had locked up and pointed it toward something that was truly meaningful to me. 

When I started minting one a day, taking a random output and sending it to a random address, people didn’t know what was going on. They were mad at me, saying: “why don’t you just give that to me, it would change my family’s life.” It was very intense. 

And it was also a lot of work as I had this whole plan to try to highlight the insanity of the hype by making outputs available for anyone with an Ethereum address, with all the funds going to the food bank. I don’t know how I’ll ever be able to funnel that much into something I think is good again.

I followed up Ringers with The Eternal Pump (2021), named after the well-known crypto meme, “respect the pump.” Back then on Art Blocks, when you did a curated drop, you would then do a Playground project maybe a month later as a service to fans and collectors. The Eternal Pump (2021) was basically saying: “this is all crazy, but you have to respect the pump.” Unfortunately, despite selling out immediately, the drop felt like a nightmare because of all the bots, FOMO, and speculation.  

I felt like something had broken, that something had changed fundamentally and that the monster was out of the bag. The era of generative art that I was familiar with and comfortable with was over.

That prompted me to turn The Eternal Pump into a kind of fellowship. The Pump collectors are a really interesting and tight-knit group — we often have dinner together and hang out in different cities when we happen to be there at the same time.

‍JB: In 2021, a lot of people entered the NFT space to make a bunch of money. But I think you had a more innocent or logical perspective that was just excited about the potential of NFTs and art on the blockchain.  

‍DC: I think I was in the right place at the right time. And then when stuff went crazy, people were looking for people to talk to about it. That helped my stature because people came to me knowing that I had some experience and had likely reflected on things. 

The overfinancialization of anything is bad. In that situation, the artist becomes no more than a tool, which is why royalties are so core — if you take away royalties, you lose so much. One of the reasons I think the Moholy-Nagy estate and others are interested in NFTs is because royalties are a means of supporting their artists and their legacies. 

‍JB: Whose work are you excited about right now? I’ve always wondered what you think about die with the most likes, who has done some parody Ringers and things like that.

‍DC: I’m a huge fan of their work and I’ve been fortunate enough to meet them in person. I’ve also tried to get their work a bunch of times but I’ve always got sniped. I have Ix Shells’s first NFT ever. I have three works by Auriea Harvey, whom I’ve been a huge fan of for years. I used to play her video games. 

I own three Ringers myself and hopefully one day I’ll get a fourth so I can make a grid.