In this episode of the podcast, Steve and Katie are in conversation with Jason Bailey, the founder of the Artnome blog and host of the Dank Rares blockchain art podcast about technology and fine art. With a background in art and tech, Jason is one of the foremost authorities on art and technology. The conversation with Jason is wide-ranging from blockchain, provenance, smart contracts, digital art, cryptocurrency, blockchain-driven auctions, privacy, and generative art.
Resources: https://www.artnome.com/ http://dankrares.com/ https://www.christies.com/exhibitions/2018/art-and-tech-summit-exploring-blockchain https://superrare.co/ https://www.cryptokitties.co/ https://portion.io/ https://hyperallergic.com/468060/christies-sells-ai-generated-art-for-432500-as-controversy-swirls-over-creators-use-of-copied-code/ http://rarepepedirectory.com/ https://www.theparisreview.org/blog/2018/01/23/much-pepe-scenes-first-rare-digital-art-auction/
Steve Schindler: Hi, I’m Steve Schindler.
Katie Wilson-Milne: I’m Katie Wilson-Milne.
Steve Schindler: Welcome to the Art Law Podcast, a monthly podcast exploring the places where art intersects with and interferes with the law.
Katie Wilson-Milne: And vice versa. The Art Law Podcast is sponsored by the law firm of Schindler Cohen & Hochman LLP, a premier litigation and art law boutique in New York City.
Jason Bailey: I think a lot of times I go to these conferences and people are like, "someday, we will buy machine learning artwork,” and “someday, we will be able to do it on the blockchain,” or “someday, there’ll be this secondary market commission.” It's all already happening.
Steve Schindler: So, we are thrilled to have Jason Bailey on the podcast with us today. Jason, who is the host of the Dank Rares blockchain art podcast, describes himself as an artist turned analytics nerd. He is the founder of the Artnome blog and the creator of a database of the -- in his words, world's best known artists. With an MFA from Massachusetts College of Art and Design and 20 years of experience in the tech world, he is now at the forefront of thinking about blockchain in art as well as artificial intelligence in art. And, I started to notice, Jason, that you are a speaker at virtually every conference about blockchain and art. So, welcome to the podcast. Jason, tell us a little bit about yourself and how do you got so interested in art and tech and came to this work.
Jason Bailey: Great, yeah, thanks Steven and Katie for having me on the podcast. So, I actually -- I grew up in a family of engineers where my dad was an engineer and my older brother became an engineer and my younger brother became an engineer. and, I decided from a very young age that I wanted to be an artist. So, I was definitely the black sheep within my family. Pretty much every dinner conversation was about new technology and, you know, math and all these things that really weren't of interest to me, but I ended up going to school for painting and sculpture and, you know, essentially studio art and double majoring in art history. And, like a lot of students in art, after I graduated, I realized that I had to like pay for food and pay rent and I got married pretty young. And, there weren't a lot of opportunities out there to sort of be a working artist, and so I caved in and went to the family trade and worked in tech for 20 years or so and took a break midway through -- actually not really a break, but went to school nights and get my MFA because I sort of missed being around artists. But, it's always just sort of been natural for me to have one foot in tech and one foot in art.
Steve Schindler: So, tell us a little bit about Artnome then. How did you come to create it?
Jason Bailey: Yeah. So, about 3 or 4 years ago, I read this book called Provenance, and in the book, they talked about — there was a famous pair of folks in England, one was the forger and one was sort of the business man and they were kind of, you know, not only forging paintings, but forging the provenance, and that really bothered me. So, as someone who works with really big companies and massive -- I mean people talk about big data, but, you know, I worked in a data analytics company for, you know, 3 or 4 years and our customers were like Facebook and companies on that scale that were dealing with massive amounts of data real-time. So, I had better around these really big databases and analytic systems and I knew that the art market you know is a pretty massive market, right? So, I think people throw around, the number around 65 billion dollars or something like that, I don't know if that was last year or the year before. So, I was trying to reconcile how is it that we don't have a database that — where is the database that would prevent all of this forgery, right? You know conservatively, the numbers that I've heard are like 15 to 20% maybe of works that are on the market and in museums are either forged or misattributed. And, I think those are obviously the nature of that statistic, it's hard to know for sure. But, I think that's on a lower side of what I've heard. So, to be honest, I was actually a little depressed when I was reading that book. And to me, art and art history is the closest thing that I have to like religion. It's been a big part of my life since I was a little kid and these artists are my heroes. So, to at a later stage in life, to read that the art historical record is so vulnerable, kind of drove me crazy. So, I'd emailed Harvard and Princeton and the Smithsonian and the Getty Institute, and all of these places asking, “where is the database that lists out the complete works by all these important artists?” There has to be one, right? I mean for historical purposes, but also for this market, this massive market. Then, every single one of them responded sadly, no such database exists, and if you find one, we'd be happy to pay for it and like, but we’re doubtful that anyone can solve this problem. And, my wife suggested that I actually try to solve it. you know like even if I have the start really small, the description that I used is that's almost like loosening the lid of a jar where maybe I can’t take the lid completely off, but if I can draw some attention to the problem and show that one person can do a lot to help move things forward in that direction, then that will be good. So, yeah, you know the quicker version of the story is my brother helped me build out a book scanner and these catalogue raisonnés that are pretty hard to find, very expensive. You know, the example I use is the Picasso catalogue raisonné, until fairly recently was about 200,000 dollars. I think you can get a 20,000 dollar version now of the Zervos. So, these aren’t easy things to get to, and I started, you know, essentially pulling the factual information from them. And, it was you know -- we could talk about this too, but it is my understanding that facts are necessarily something you can copyright, especially if they are presented in chronological order, and pretty immediately started discovering some amazing things, right. So, I'll stop in a second here, but like even basic information like what's the average size of a Cezanne painting or like you know there are at least things that I kind of chuckle at companies that are starting to launch that call themselves like art analytics companies because without the data, you really can't say much or do much, right. So, I feel like this dataset that I’ve built was really giving me insights that a lot of people have just never had access to it before.
Katie Wilson-Milne: So, Jason, your first step in this database was you located a bunch of catalogue raisonnés and you went to libraries that had them and you scanned them?
Jason Bailey: Yes. So, I bought some. So, it is really like crawl, walk, run. I bought some and realized that it's pretty expensive. I wasn't going to be able to buy them all, right. I just didn't have the means for it. And then, my MFA, just by luck, is from the only state run art school in the country, so Massachusetts College of Art and Design. And, they’re a super, super liberal library. So, most of these libraries, if they have a catalogue raisonné, and most don't, wouldn't let you take them home. But, this is art school, right. So, it's a state run liberal art school. So, they are like, "Yeah, just take." You know, as an alumn, I was able to take them out. So, I started there and built my collection from there. And then, I think I paid like a membership fee, at BC. They have a beautiful, beautiful art library, the Bapst Art Library, it’s one of the nicest ones I've been into. And, I was able to kind of grow the dataset from there.
Katie Wilson-Milne: So, you scanned the pages of the catalogue raisonné, and then, you by hand, input all the data that's there, right, because you are not using these scanned images other than as inputs, right?
Jason Bailey: Right. So, they are just inputs. So, I build sort of a whole data work pipeline. So, I had them scanned and I am able to use character recognition for some, but I wasn't super happy with the quality there. So, I ended up offshoring some of it to some folks that sort of do the transcription and into a database and then I kind of clean up that database using some custom tools.
Katie Wilson-Milne: So, how far along are you?
Jason Bailey: Yeah, yeah. So, I got to — I'm somewhere between 40 and 50 artists, which may not sound that big, but I'm pretty confident that it's the largest database. You know, I've been saying that for a while and no one’s corrected me yet.
Katie Wilson-Milne: And, you are starting with you know the world's most famous recognized artists, I assume like, Picasso, Cezanne, Matisse, you just mentioned.
Jason Bailey: Yeah, yeah. I don't have Picasso yet. That's one of my white whales, but it's definitely some of the better-known ones because those are more accessible, but also there is a great overlap between the artists who sell for the most in the art market and the artists that have catalogue raisonnés, right. I don't think that's coincidence.
Katie Wilson-Milne: Right.
Jason Bailey: So, the artists that we care about the most actually tend to be the ones that are best documented. So, yeah, it's a lot of pretty well-known names and, you know, as I got further in, I shifted towards artists that were a little bit earlier on because the image rights gets less sticky I think after they've been passed away, what is it 70 years, or something like that?
Katie Wilson-Milne: In the US. Yeah.
Jason Bailey: Yeah. So, I thought, you know maybe it make sense to shift to that side. You know, it's been a while since I have gone back in and added new artists because I eventually realized that I've just been run out of time and money and nobody knew what I was doing, right. So, I was just like spending literally thousands and thousands, you know, of dollars and hours preparing this data by myself. And you know, if I had gotten hit by a bus or whatever, no one would know about it, right.
Steve Schindler: Right. And, this is -- this sounds like the ultimate centralized database, and we're going to talk about decentralized databases in little bit. But, I'm curious what's the end game for you and the databases. Is there a way to monetize it or what is your goal with it?
Jason Bailey: Yeah, so I shift back and forth on it. My goal is actually to help solve this — I see this as a major problem. Even more so than a business opportunity, I see it as a problem I want to try to help solve in my lifetime, right. So, art has given me a lot. My identity and you know got me into school and it's gotten me my jobs, and I'm really big into art history and I'm greatly saddened that the historical record is what it is. So, the goal is to really do my part to help trigger like an art analytics revolution. I was part of the early basketball data revolution, so I've kind of seen this play out. And, I think when you get…
Steve Schindler: It’s the money ball of art.
Jason Bailey: It really is. And, when I say that people think that maybe I’m more focused on money than I am, it's not, you know -- I don't know yet that I've cracked the code on a business model for this, but I know from the basketball analytics revolution that when you have a good clean dataset and it's publicly available, very smart people from, especially from universities start to flock to that data and find new things that no one could have discovered before, right. And, so with basketball, you know, there were these conferences every year at MIT and you know the league was slow to adopt it, as I assume the art world will be slow to adopt sort of an analytics revolution. But, they have now, I think, similar things could happen in the art world.
Katie Wilson-Milne: Do you think Jason that you will have support and buy-in from — the key players, in my mind are artists’ estates, catalogue raisonné associations? You know, I guess on a utilization level, auction houses -- I mean have you had feedback from any one critical or otherwise, like an artist’s estate for example, about what you are doing and how you are using the materials they put together?
Jason Bailey: Yeah, I've been very slow. So, my end game would definitely be to make all of this data widely available, but I don't know enough about the legality of it. So, I have been somewhat timid about how I share, but I think I certainly can share insights if not data and I need to, you know, to dig deeper on that. In the beginning of this process, not that I would say I am very well connected necessarily now, but as you mentioned, I get invited to a lot of these presentations and these conferences. and, I've to say it's been nothing but positive. I was pretty scared and paranoid at the beginning that like you know maybe if I let this story off, people are going to come find me and I know there is like wealthy collectors out there and stuff. But, everybody, literally everyone that I've talked to has sort of embraced this and I think people realized that it's something that we need, you know, in order to take care of these important artworks, right. So, it's been all positive so far.
Katie Wilson-Milne: Tell us a little bit about how your blog and your podcast come out of this interest and how that relates to what you are doing now.
Jason Bailey: Sure, yeah. So, after 3 years of spending all my money and time scanning books and building a database and realizing that no one else knew about it, I knew that I would have to launch some sort of platform to talk about this, right. And, Artnome really came directly out of that. So, my first few articles were about trying to use some of this data to make predictions in the auction market and what the new statistics around artists could look like and why that would be important, and pretty quickly, within an article or two, people started paying attention. So, FiveThirtyEight’s a very well known journalist, it would be like a journalism outlet for lack of a better word and they are very analytically bent and they came and did a story on me at my house and interviewed me. And, that kind of like launched me to where a lot of people became more aware of Artnome. So, you know I wrote exclusively about data and art analytics for a few months, but then, you know, I have other interests at the intersection of art and tech and I found that as I wrote about these different things, there really is an audience for the intersection of art and tech, and not a lot of people writing about it who have both an art background and a technology background. I mean there are definitely people writing about it, but I think it's rare that you get someone with a background in both, which led to this blockchain article I wrote about 6 or 7 months into launching Artnome. So, a friend of mine who I had done an article on machine learning for prediction in auction houses, we had dinner and he said, "You know, Jason, you should really look at blockchain. I think, you know given your interest in provenance and tech and art, you know this would be something you would be interested in." And, my reference to blockchain was that it was like the nerdy guys that live in their parents' basements who are like mining bitcoin and stuff like that. I am like, "I don't know if that's a thing. Maybe you don't understand me because I don't feel like that's my thing or whatever." But then, I did about half a day's research and wrote an article called “The Blockchain Art Market is Here.” And, like 2 days later, my email inbox was just loaded. I kind of just accidentally stumbled into the top search rank in Google for blockchain and art overnight.
Steve Schindler: Oh, that's great. And, I read that article. It's really good.
Katie Wilson-Milne: So, how does that -- how did you transition that to the podcast are you still sort of equally focused on both the database, the blog, and the podcast?
Jason Bailey: Yeah. So, I am an expert in not understanding things and admitting it. So, you know it just came along with being the least technical guy in a family of engineers. So, after writing that post about blockchain, I got so many questions that I couldn't answer that I had to find real experts, right. People thought I was an expert and I wasn't. So, I started going to conferences and talking to people and fell in love with the community. It turns out the people in their parents' basement mining bitcoin are actually pretty great people. And, I made friends and, you know, obviously joking around. They are not all like that, but I started recording my conversations with them so that I could write articles. And, I realized transcribing these long interviews that I was having with all these experts who were really generous with their time talking about how this stuff could work rather than transcribe them into super long transcripts that I had all the materials I needed to launch a podcast. So, I launched the Dank Rares Podcast thinking maybe some small niche of people would be into it, and actually, it grew pretty quickly.
Steve Schindler: And so, just tell us what -- for everyone who is listening that doesn't know what does Dank Rares actually mean?
Jason Bailey: Yeah. So, that was an interesting one too. I almost didn't call it that because for a lot of people, “dank” is a generational thing. So, I think if you grew up in the 60s, 70s, or 80s, the term “dank” was usually in reference to like marijuana, you know like pungent marijuana.
Steve Schindler: Right. That's the world I grew up in.
Jason Bailey: Yeah, yeah. And, I kind of -- you know I grew up to a fair degree in that world, too, but the millennials often use “dank” just as a substitute for cool. It's just like, "Aah, it's cool."
Katie Wilson-Milne: I don't do that.
Steve Schindler: Other millennials.
Katie Wilson-Milne: I am like the oldest millennial possible, so you know that's why.
Jason Bailey: Yeah. I’m like I considered myself an honorary millennial. So, yeah, you know “dank,” as I understand it, you know means cool and then rare -- it's sort of slang for rare digital art from the early days like the Rare Pepe days and things like that. So, I was very surprised to see that the domain name was available and I knew it would resonate you know where Artnome maybe is more for art type folks. You know, the art market people that are little bit maybe more associated with that. I saw Dank Rares as more resonating with the folks, a younger crowd that were actually like, you know, producing artworks on the blockchain. So, the name seemed to fit. I do get some strange questions about it, but everybody remembers the name, no one -- it's memorable.
Steve Schindler: Right. That's how I found you originally was through the podcast, listening to that.
Katie Wilson-Milne: What was the Rare Pepe example that you talked about? Can you explain that for our listeners?
Jason Bailey: Yeah, yeah. A lot of times, people completely skip over Rare Pepe when they want to talk about the history of digital art and the blockchain and it's understandable, because so Pepe the Frog was a meme that everybody was using — basically, you could use it to say and do whatever you wanted as it was just like -- think of it like that a modern day Mickey Mouse, right. And, some people would you know use it to make jokes or like political statements or this that and the other but, it got to a point where the alt-right pretty publicly was known to have kind of commandeered Pepe. So, it becomes a hard -- it's a hard topic to talk about, because if you are going to explain blockchain, which just confuses people enough as it is and now you have to explain how this -- what most people think is an alt-right symbol, there is a lot there to unpack, right. So, what a lot of mainstream journalists do is just skip over the fact that the Rare Pepe Wallet existed, but it's important to know that part, because Rare Pepe Wallet really is where a lot of these ideas for something like CryptoKitties or CryptoPunks. All of those spanned out of the Rare Pepe Wallet. So, quick definition of Rare Pepe Wallet, one of the jokes around Rare Pepe you know is that it's rare, right. So, it's digital and it's infinitely reproducible, but it was like an ironic thing to say that like, "Oh, this is the rarest Pepe or this one is more rare than that one." So, Joe Looney and a few of these other folks that helped build out the initial Rare Pepe Wallet made the thought connection that using digital scarcity in things like blockchain, you actually could make these digital things rare, which was like even more tongue and cheek. So, they built this market place and people could submit their own versions and it was open to anyone, and eventually, it took off when the cryptocurrency boom happened last year and the prices went way, way, way up. A lot of these folks that had money -- there was only so many things you could buy using cryptocurrency and the market for Rare Pepes took off and there was famously this Homer Simpson Pepe that sold for like 50,000 dollars or something like that that a lot of…
Katie Wilson-Milne: All in bitcoin, or like what is the denomination?
Steve Schindler: Yeah. So, Rare Pepe is all bitcoin. These days, a lot of the market places that people talk about and things like CryptoKitties and CryptoPunks in some of the art market places are Ethereum-based, but a lot of the original projects were based on bitcoin.
Steve Schindler: Maybe, this is a good time to segue into blockchain and its relationship with art. We recently had on the podcast Nanne Dekking who you know, who is the CEO of a company called Artory that is using blockchain technology in connection with the improvement of provenance and title records and the way that we were just talking about. So, our listeners know something about blockchain, but we’re all in some ways beginners here. So, why don't you just give us a kind of a basic definition of blockchain and how it works and then we can segue onto its connection to the art world?
Jason Bailey: Yeah. So, I usually keep the definition pretty high level and simple and then I keep my discussion points to the aspects of the blockchain that I think are particularly useful to the art world. So, I don't -- I'm not like a database aficionado per se, but you know blockchain is essentially a database. People sometimes call it a distributed ledger and the idea is that it creates an immutable record, right. It's a decentralized immutable record where you can add things to it and you can’t go back and change that record, and everybody will be able to see that that's permanently part of the record. So, that in and of itself like not super exciting to me, but the three areas where it clicked for me the day that I wrote my very first article were that it creates this potentially this better opportunity for provenance. So, I what I'd heard when I started scanning all these books and trying to build this database is that the reason -- one of the reasons why we don't have better provenance today is because collectors don't necessarily want to divulge their personal information, but they wouldn't necessarily mind divulging information about the artworks themselves and the sales, they just don't want to be tied to it, right. So, the blockchain allows for that layer of anonymity of the collector while exposing the results publicly in a way that's timestamped to the blockchain itself. So, that clicked for me pretty early and I think Nanne is a great example and Artory is a great example of that in action. You know I was sort of just speculating that that could happen over a year ago when I wrote the first article, and it's great to see Christie's and the Ebsworth Collection, you know, that this is something that actually can happen. So, that was one leg of the stool. Another thing, you know, having told you my background about graduating from art school and not being able to find any way to make a living off of it, that's another thing that I care a lot about, too. And, I saw that there was this opportunity to use something that people refer to as “digital scarcity” to make it possible for digital artists to buy, sell, and trade their work online. So, the premise without getting too technical is that the same thing that makes bitcoin valuable because you can prove that you know you've got one through the blockchain and you can destroy it and lose it and trade it. That idea of digital scarcity can be transferred over to digital art as well. and, you know around the time that I had first written that blog post, the Rare Pepe Wallet was sort of the biggest example of that but since then, CryptoKitties and CryptoPunks and things like that have launched and kind of proven out this idea that people are willing to accept that having the token associated with the artwork creates a viable market right for these artists and cuts out this idea of a gallery because you could kind of do it on your own and sell directly. So, that was the second really kind of appealing aspect of blockchain for art and then the third one, and I have been less into this in the last 6 to 9 months, but this idea of fractional ownership, right so that you could take a work — almost like an art stock market. There are a few variations on it, but you could take a work that you and I can’t necessarily afford and break it up into these pieces and prove that you owned it across a blockchain. And, in some cases, people are literally breaking up the artwork itself, so Kenny Scharf, I think, in Art Basel Miami did a piece where people kind of ran and grab the chunk of it and I guess each piece, physical piece was supposed to be tied back to the blockchain. And then, Eve Sussman with Snark.art, is a project that’s sort of interesting, she took her “89 Seconds at Alcazar” video project and broke that up so that folks could sort of — expanded, I guess, the number of people that would be able to take part in owning her work. So, I mean I support experimentation. I just -- I wonder with physical artworks, if you are doing fractional ownership across a thousands of people, do these works just end up in free ports where no one can see them? Like, you know I don't have the answers to those things yet, but I do have some concern. So, yeah, I think those are the 3 areas that I look at.
Katie Wilson-Milne: I think we want to ask you a bunch of questions about those areas. But, first, before we move more into digital art and generative art which we want to talk about, and I see as sort of a separate category from some of these blockchain questions in terms of technology and art. I mean, is the major contribution of blockchain to the art world that it is replacing however imperfect a system of legal relationships that previously were you know captured in written contracts or verbal contracts that there is sort of a shift in the legal relationships that are occurring and a formalization of those relationships? I mean, because when we -- when Steve and I think about blockchain in the art world, we are waffled between being confused, but impressed and confused and concerned and a lot…
Steve Schindler: And skeptical to some extent.
Katie Wilson-Milne: .. and skeptical yeah. And, a lot of that is because we don't -- it's not always clear how this fits into how the law recognizes, how parties relate to each other around the creation of art, around the purchase of art you know and other things in the artworld like lending. So, I don't know what are your thoughts on that sort of relationship piece.
Jason Bailey: So, I'm old enough that I remember when the internet was truly wide west in early days and I think blockchain over the last year, and it’s starting to get much more regulated, but over the last year, blockchain has been fairly wild west. So, you are not alone in your confusion around its potential and its usage. I think we were in a sort of a rich period, so I kind of like uncertainty, I like the early stages when no one really knows what the full potential of a new technology is. And, I think what we saw in the last year and half is a lot of people trying to see what you can do with it and a lot of enthusiasm around blockchain and to your point about the legalities or streamlining the legalities or potentially taking out, you know, the word “middleman” is used to describe just about everyone, right. So, it's like lawyers or galleries or auction houses, right. And, there is this idea, this ethos with blockchain that decentralization is a good thing, and, if we can have transactions where we don't have to you know write up a fresh contract each time and we can go around, bitcoin and banking is the prime example that people would use, right. So, you don’t have like credit cards take out a fee or a cut, but if you can pay people peer-to-peer using something like bitcoin, you don't have to pay the middle person, right. And, I think there are some people who thought that you could do that with art. So, maybe it's a good time to talk about smart contracts. So, a lot of the innovation around blockchain and art is using these things called smart contracts, right where you can write up a set of rules that govern how a DApp, a decentralized application, works moving forward. And, I'm not a legal expert or technical enough to write my own smart contracts, but the idea is that once you set these things in place, everyone has to follow them and there is no longer someone that manages the process. Let me give you a real example, because I know that sounds kind of an abstract. So, I bought a CryptoKitty. you know. early on just because it was the easiest thing for me to buy and I wanted to understand how well this stuff work, right. So, I bought 2 or 3 of them. I wanted to give one to my niece, because I thought it will be cute, you know. And it was this CryptoKitty called Limoncello, or whatever. And, my younger brother said he set up a wallet. He is a technical guy, sent me his address, and I sent it to that address, but he never received it, right. So, the thing went nowhere. Who knows where it went, right? Maybe ..
Steve Schindler: If you have a decentralized database, ..
Katie Wilson-Milne: Yeah.
Steve Schindler: ..there is no one to call, right? It's ..
Jason Bailey: That's exactly what I'm getting at, right. So, it wasn't like I couldn call CryptoKitties and be like, "Hey, Limoncello, like disappeared. Can we undo or anything like that,” right? And, it's the nature of the smart contract that all that stuff kind of just gets governed separate from the institution, right. So, yeah, Limoncello’s, I think someday if someone opens up a new account and gets that, whatever number I sent it to, they may actually just have a cat waiting for them. But, I don't know what the odds of that are.
Katie Wilson-Milne: But, here's the thing you know about smart contracts, which is really a question I think for lawyers about blockchain's potential to govern relationships between people rather than just create a ledger, you know, for a particular group. Somebody is still deciding what those terms are going to be, right? A smart contract is only as good as some mutual understanding about what the terms should be, which have to be still a human negotiation. And then, an ability for people to come to that smart contract and understand fully what they are agreeing to and be okay with it. In the art world, you know unlike music and book publishing and other forms of creative expression, every interaction is bespoke in a way that the music licensing isn't and book publishing really isn't. So, I don’t know. I have questions about how that's going to lay on to the art world when there is really no history of even standardized contracts for a lot of transactions.
Jason Bailey: Yeah. So, a few things there. One of the things that I -- when I talk about the art world in particular, you know databases have been around for a long time, right. And, one of the questions that I have is, is the opacity, the lack of transparency in the art market, a technology problem or is it a cultural issue, right? Because if it's a cultural issue, you know I mean no one embraced previous databases, right. And, if it's a cultural issue, blockchain doesn't necessarily solve that. I do think that it was an opportunity the mania or enthusiasm around blockchain and art that we saw last year was an opportunity to get the need to bring more transparency into the news and sort of into the forefront, but I myself wonder you know what's the magic about blockchain that you couldn't have accomplished with a regular database in the past. And, I do think that there is some cultural friction around that.
Katie Wilson-Milne: Yeah, I agree.
Steve Schindler: Yeah, I think we both agree. And, you know there are a lot of people in the art market who, I mean everyone talks about wanting greater transparency, but I think the truth is there are a lot of intermediaries whose businesses to some extent is augmented by the lack of transparency.
Katie Wilson-Milne: Right. And, that's how they make money.
Steve Schindler: That's how they make money. It is because they have information that other people don't and thus, you have to go to them to get it. So, I think there is that tension there as I think I saw you know looking at the room at the conference. I think one of the — part of the title of the conference was “Is there a consensus in the art market?” And, I saw lot of people sort of seemingly shaking their heads, at least figuratively.
Jason Bailey: Yeah, I mean from my perspective, I am maybe less involved in the art market and more excited for the historical implications of it. So, I, you know again going back to having read that book and realizing how susceptible our art historical record is just broadly. I feel like it's something that needed to be addressed, but I understand it's important to understand friction points around collecting. And, the other note to your point about people not understand the smart contracts, you are absolutely right. So, in a lot of cases, at the height of sort of the Blockchain boom last year, people were just cutting and pasting and reusing smart contracts. You know I read somewhere, I forget the exact numbers, but there is like some crazy overlap in the way the contracts were laid out across all these different applications because most people don't have you know a combination of the legal expertise and the technical expertise that it would take to sort of do something from scratch with one of these. But, I think for those who do understand those things and do have the right teams together, you can do some pretty interesting work.
Steve Schindler: I’m also intrigued by one of the things that blockchain and smart contracts may have to offer, which is to try to solve a problem that we have been really resting with here in the United States which is artist resale royalties. A resale royalty, which is something that exists in most European countries allows for an artist who creates a work, who sometimes historically has not benefited by the rise and value of that work to receive a royalty when that work is sold to subsequent buyers. There’s pros and cons to this and a lot has been written about it, but I think the lot of people who support artists feel strongly that it's a worthy idea. And, it seems to me that blockchain has the ability to contractually or at least conveniently allow that to happen here.
Jason Bailey: So, part of what's fun about what's going on in blockchain, I mean granted it is not a huge scale, but a lot of things that people sort of speculate like is it possible, isn't it possible actually already are happening on a small scale in really cool ways. So, SuperRare.co, it’s either .co or .com. but, they are one of the many art market places for digital art that's build on the Ethereum blockchain and they baked into their smart contract that something like 10% of every secondary sale goes to the artist, right. So, in this community, you know I remember there was a moment, I want to say it was last April where I bought — I was one of the first people to buy on SuperRare. In April of last year, I bought a machine learning artwork on the blockchain and there was a secondary, you know I ended up selling one of them. I bought a bunch, and one of them, there was like a secondary commission or resale commission that went over. So, all of these things are actually happening. I think a lot of times I go to these conferences and people are like, “Someday, there will be a secondary,” you know or, “Someday we will buy machine learning artwork,” and “Someday, we will be able to do it on the blockchain,” or “Someday it will be this secondary market commission." It's all already happening. And, what's interesting is kind of the -- and granted on a small scale, but what's interesting is that it shifts the dynamic. You know, I think the art world has reputation about a small number of people spending a lot of money on a small number of dead artists, right. I mean that's not super fair, but it’s — I mean it's to some degree, right.
Katie Wilson-Milne: And living artists.
Steve Schindler: Yeah.
Jason Bailey: And, I think what we are seeing on some of these digital art market places is that the dynamic is shifting and people are more likely to resell if they know that 10% is going to go to the artist. Like it’s shifting from just “I want to own these objects” to “I want to support these artists,” right. If I buy and sell a work, instead of thinking “Is the artist is going to be mad at me that I sold this because there is too much volatility?” they are going to be happy because they are getting 10%, right. And now, somebody else can participate in it, too. So, yeah, I think that, as a person who went to art school and wanted to be an artist, I think that's a very important thing to bake in and if we can't get it at like a government level, I think what we are seeing with these experiments, admittedly they’re like small scale, is that there are ways to bake it into the -- voluntarily bake it into the smart contract.
Katie Wilson-Milne: Do you think, and maybe this be our sort of last blockchain point of discussion, do you think it is more promising, let's say when compared with smart contracts, that the art world will move towards cryptocurrency payment? I mean is that something that seems more realistic to you, are more promising and if so, why would that be, other than you know money laundering or hiding assets?
Jason Bailey: Yeah, I don't really think about or talk about — it's just not that sexy to me the using crypto as payment, but your point about wanting more opacity, that's an example where blockchain could enhance opacity, right?
Katie Wilson-Milne: Right, right.
Jason Bailey: So, if you wanted to make payments and didn't want to disclose who you are or you just didn't want to have to go through a third party and you have the convenience of making a payment from one country to the next without necessarily having to go through the governance and the implications involved with it, yeah I could see that. It's not an area I’ve given as much thought about, but I could see it going, you know, increased usage. I think it will basically, you know, it’ll follow with increased usage in cryptocurrency in general. If we see cryptocurrency getting increasingly adopted as a primary method of payment, then I think we will also see that with the art world. But, for me, really the 2 things that keep me coming back to it and really excited about it are the potential to help improve our art historical data by adding sort of a public provenance system that we can all build on and grow from. And then, it's really exciting to me to see that there are an entire generation of digital artists who have sort of for the first time arguably a way to sell their work, when it's you know something on a screen that everybody can see and infinitely replicable in some way. So, there weren't a lot of good systems in place for them to make their work sort of ownable as it were.
Steve Schindler: So, one other topic I wanted to talk to you about was, and I know that you are a little bit involved with a company called Portion, which has to do with the sort of blockchain-driven auctions and where do you see the future of that in the art world?
Jason Bailey: Yeah, yeah. Thanks for bringing that up. Yeah, I am an advisor for Portion and also for DADA.nyc, two really interesting different companies working with art and the blockchain. So with Portion, part of thinking is that if you are under the age of maybe 25 or actually even if you are over, the art world is intimidating, right. Galleries are intimidating. Auction houses are intimidating. If you go in and you don't look like you have, you know, enough money to spend, you know people might look into you the wrong way and who knows who has money. It just, even you know my experience with the auction world has been very welcoming, but whether it's true or not, people see it as intimidating, right. And, the idea is that the next generation of collectors maybe want a new system or new process. And, I think what Portion is thinking is that anybody could become an auction house, right. So, these younger collectors if you have valuable goods, there are a certain amount of contracts and oversights that are bid into the auction process that maybe could get pushed to the blockchain. And, you know to expand upon that, the survey results that I have seen and from talking to younger folks, I also think there is a trend towards wanting a very liquid market where you can buy and sell frequently and in a more traditional art world that sort of looked down upon, right. So, you know the idea is that you are buying the art because you love the art and that you are going to hold it for generations, right. And, if you are a flicker, that's sort of a negative connotation, but I think part of what Portion’s tapping into is they are realizing, okay, there is a young generation of folks, a lot of them have money, they care about luxury goods, not just art, but luxury goods, too. They don't feel like they are welcome at the traditional you know 100, 200 year old auction houses, is there an opportunity to build a market where they can sort of buy and sell and trade being their own auction houses with lower fees, lower commissions that would appeal to them?
Steve Schindler: It sounds like they need a good lawyer and, you know, we are here to help.
Katie Wilson-Milne: That's where that excess capital should be used. Jason, let me ask you this, this is an equally common question for you and for Steve. But we haven't talked about privacy concerns yet. And, you know Europe is taking a sharply different turn with how do deal with privacy in the online age than the U.S. has taken and really than the U.S. can take given our constitutional, you know, advantages or constraints. So, how would blockchain and the idea of an immutable ledger or a publicly available immutable database of information that might have personalized either actual facts or hints or something, how does that work in a sort of European GDPR construct where you have a right to be forgotten, you have a right to immediately ask for your personal information to be taken down? You know I mean, how do those map onto each other, because in some ways they do seem to be going in different directions?
Jason Bailey: I mean one thing that comes to mind is you can separate the personal records from the other information like the Ebsworth Collection, right. So, Christie's wasn't storing the collector information on the blockchain, they kept those separate partially probably to protect their client's interest, but also partially because of — I know they themselves had gone through the GDPR process first half of last year.
Steve Schindler: Right. So, this is actually a good place maybe to segue to another topic that you’ve written on and a related topic, which is called generative art and also artificial intelligence, or AI art. Can you just explain a little bit about what is generative art?
Katie Wilson-Milne: Yeah, and how does it relate to digital art?
Jason Bailey: Yeah, yeah. So, generative art actually has a pretty rich history going back to the early 1960s and the idea, I mean different people have different definitions. I'll probably get slaughtered by my generative art friends for oversimplifying, but the idea is that you are writing a program that produces the art, right. So, digital art could be you hop in to Photoshop or Illustrator and you are using someone else's algorithms like a paintbrush to try to create images, but with generative art, you are actually writing code, and if you want some fancy, algorithms, right, that are producing the artwork, right. So, that's high level what generative art is, but there are a lot of things that fit under that umbrella, including AI and machine learning art. So, when most people think of the term generative art, you usually think of something that sort of geometric and it looks like it's the product of algorithms, right. There is a lot of repetition and complexity and often it's you know 99% of time it's something that's fairly abstract with no recognizable visual elements. And there is some really, really beautiful, beautiful work, generative artwork being done, and then in the last 2 or 3 years, people have been doing more with machine learning. And, you know that is also generative artwork, but it doesn't look like what we normally think of, which would be sort of these complex geometric-type works. And, what's going on with the machine learning art, so people call it AI art because it's just easier way to talk about it. But, the closer you get to the artists, the more you realize they don't really like description. So, it's you know technically probably machine learning art, not just AI art, you know, that would be a more descriptive way of talking about it. And then, when you get into the tools that most people are using today, it's actually these generative adversarial networks. So, they are neural networks, and I'll do my best to do a sort of a quick description, because this is the stuff that people are seeing in the press today, right. So, the work from Obvious that sold at Christie's end of last summer; the work from my friend Mario Klingemann that's for sale at Sotheby's right now. So, they are using generative adversarial networks and it's essentially two networks and one is — so, people kind of describe them as an expert in a forger sometimes as a good way to help people understand how they work. So, the expert's job is to look at everything the forger makes and try to figure out if it belongs in the dataset or if it belongs outside the dataset, right. So, let's use the example of historical portraits. If the forger sees you know 10 historical paintings or whatever, it tries to produce something that looks like those, but isn't those that it can fool the expert with. And, with just 10, you know it's probably not going to do a very good job and the expert can tell right away, "Okay, that doesn't look like the 10, that's a forgery." But, if you feed that forger network, say you know 10,000 or 70,000 historical portraits, it gets pretty good, right, and it can produce something that we've never seen before, based on what is learned from all of these training images that it’s received. And, the expert network can't tell the difference anymore, right. So, there’s this battle that goes back and forth had a high number occurrences until you fool the expert. Now, the stuff that people are finding really interesting, these really weird surreal kind of images like weird, trippy you know nudes and landscapes. What's happening is sometimes the expert is not that good, right. So, it start to let things through that it thinks are, you know, should be part of the training dataset or the historical portraits or whatever images we are using to train, but they are off, right, in sort of interesting ways. And, that's where some of the artistry comes in for the GAN related artwork.
Steve Schindler: Ao, you mentioned the work that was sold at Christie's by the Obvious collective, I think and -- can you just explain what that was and the controversy around it?
Jason Bailey: There is an increasing problem that we're going to see — this is probably a really interesting legal-related question. So, in software, art aside, in software we have had these sort of open source licenses, right, and there is all kinds of different licensing schemes and there've been legal battles for decades around how software gets licensed and you know what you can use and what you can't use. Now, would you overlap art on top of that and those headaches become part of art law too, right. So, we have got artists that are building all these licenses. So, Robbie Barratt, who is a good friend of mine, who is a 19-year-old kid from West Virginia who just happened to build out these brilliant generative adversarial networks and had written some of the code, but he too was using code written by someone else, right. So, there is Ian Goodfellow who is actually the first, who came up with GANs, or is generally credited anyway. And then, I think it's Soumith Chintala who had written the python implementation that Robbie expanded upon. But, it's enough to say that someone came up with a concept, someone else wrote some code, Robbie used that code and added that code, but also came up with the idea of using the historical portraits and ran them through the GAN. And then, he open sourced his part of the code, right so that people could use it. So, this French collective, 3 guys in their 20s or whatever, came across Robbie's code and were bugging them pretty regularly saying like, "Hey Robbie, can you like change it for our use? If you do this and can you give us a little bit more of it?" Robbie said, "Yeah, yeah, I guess so, I guess so." And then, the work they created was not all that different from what Robbie had created. So, they used this open source code. They trained it with historical portraits that were pulled from the same site that he had pulled from, right. So, it was pretty duplicative of the work Robbie had already done. And then, immediately started marketing it and trying to sell it. So, I mean — actually, I interviewed, I did the longest I think interview with the guys from Obvious, because they were pretty upset around the time that it was being sold at Christie's because the community was really mad at them, the machine learning and arts community because they are like, "Hey, you kind off swooped in, you took this kid's code, you didn't really do anything new or interesting. You'd even do a different project. You used all the same training images, right and you know, you did the minimum to adjust it and then you spent all your time essentially selling his work under your name." Now, from a legal perspective, if they are using open source software, I guess there is nothing wrong there, but I think legal and ethical aren't always the same definition, right. I think people were like you know, "Yeah, maybe it's legal to do this, but is it ethical to take this kid's work and you know run with it to try to sell it."
Katie Wilson-Milne: That's a big theme of our podcast, is that there is an almost constant question in the law, in general, but in the art world about the boundary between law and ethics, and it goes both ways, right. Some things seem legal and unethical and some things seem illegal, but completely normal in the art world you know almost as if they’re technicalities. But, this area you know when it really hits the courts, I think is going to be very interesting because it gets to the heart of what are copyright laws supposed to be doing and what it's all about. And, as many smart people have argued our copyright law may exist for a prior time, you know it may not exactly map onto how creative people do things today, how those relationships occur. You know I think we think there will be a really interesting confrontation between our old common law copyright system and how works are being produced today and how ownership is thought about. And, I don't know, I mean I'm curious of your thoughts. You know even we see artists who are just not thinking about ownership in the same way until they get sued or you know until somebody steals something that they made and then maybe they do. But, there is much more you know sort of a communal attitude about creation that sometimes makes me very nervous because the law has not caught up with that and they could get a trouble for that, but there does seem to be sort of a shift in terms of how maybe creators are thinking about this.
Steve Schindler: Well, the one thing we know fairly recently is that first of all the U.S. Copyright Office won’t issue a copyright unless a human being has created a work. At least in our country, and I know it's different in other places, something that's created by a mechanical process is not eligible for copyright and then the Ninth Circuit case that came down recently with the Naruto the monkey, which a lot of people followed, it was somewhat comical in some ways, but the Ninth Circuit Court of Appeals has held that unless a work is created by a human being, it can't be registered for a U.S. copyright. So, obviously, this is going to either have to evolve or we're just going to have a different concept of what ownership means.
Katie Wilson-Milne: Right. I mean as you said, Jason, the law's relationship with software and copyright has been complicated for a very long time, and now, it just sort of looks like a part of the art world is sweeping into that sort of software bucket in a way that, you know, is probably inappropriate for artists. They don't think of their work as being like software, right. They probably think of it in a more personal sort of creative bespoke way. So, those are some of the legal issues that we think about in addition to who is really the owner, who is the rights holder to this work, is it the person who created the software, is it the person who used the software in a unique way, can a person even claim they are the artist if they just sort of feeding stuff into an algorithm and they are the only one who prints it out. So, what are your thoughts on all of this?
Jason Bailey: Yeah, yeah. No, it's great. You know one thought that comes to mind is that the community almost self polices, so I do think there is an interesting and a necessary evolution of legal that needs to take place, but if you look at what happened with Robbie, the pressure was enormous on the Obvious folks to clarify. So, a few things — they early on had said that the machine was doing all the work and their story was the first big machine learning art story that hit the world, I mean it was in every single you know main screen media and it reinforce the idea that the machines are making the art, which isn't true. And, even they know it's not true and they admitted that in the interview, but you know in early days, they thought that would help them get some attention. And, at the time they didn’t realize that they were going to become the biggest story for a few weeks and like in every major — and so, when I interviewed them, they were like, "We were just some guys in our young 20s playing around with this code, seeing where we could go with it, we had no idea that Christie’s” -- how would you know? “We didn't know that christie's was going to sell it for 450,000 dollars, right?” So, we were kind of just saying stuff off that as you would when you just -- as you would when you don't think it's going to be a big deal, but then all the mainstream media came out and said, "Well, you know the machines are making the artwork,” when in reality there is a bunch of code in work that has to be done to produce the artwork and they didn't do much of that coding, they borrowed it and it undercut the other artists, right, because now the whole world thinks the machines are doing the work and all these machine learning artists who have been working on this for a long time, and some of them are super talented. The public perception is that the machines do the work, right. And, there is this stage where after you trained the model, it's doing sort of like a black box and it's doing things that you don't — no one really fully understands. It is my understanding of it. But, that's far from the machine making its own artwork. There is, you know, a bunch of coding that has to be done with a bunch of curation. Everyone needs to shift away from worrying about whether or not machines are taking artist's jobs and that the machine is like -- those claims have sort of been overstated. In people focusing on that, what's been lost is the sort of going back to software problems and licensing. Licensing is the really interesting part, right. So, like when I use Processing, you know which was written by these two guys to write my code, which produces my art, how do you divi up credit then, and it's almost weekly that we see someone getting called out by the community, to come back to the policing part where someone is like, "Hey, you just took my code and you tweak like two things to change a color or whatever, and now, you are trying to sell it somewhere,” you know. And, even if it is open source, that just seems unethical. So, sadly I think what we're starting to see is artists are pulling away from open sourcing their code, which is -- it's especially sad because this is traditionally how artist learn to code is by using other artist's code.
Katie Wilson-Milne: So, Jason, shifting away from the artist, what is it look like as a collector or consumer of art when you, I don't know, buy a piece of generative art, I mean are you getting a file, are you getting a printout of that, so it's more like a photograph, where there could be multiple additions of the same thing or variations of the same thing? I mean what is that there, there that's been purchased? And is that satisfying to a consumer in whatever form they’re getting this art?
Jason Bailey: Yeah. It's one of my favorite topics and I think it changes from person to person. So, there’s different systems. Some people are building out systems where you can see the work at low resolution, and if you spend the money to "own it," you get access to a high resolution version and that tries to retain this idea that as a collector, you want to own something that no one else can own and that exclusivity is where the value is. But, there is a whole another school that says exposure actually drives value, right so like when I bought some of Robbie Barratt’s early nudes on this website called SuperRare and everyone has access to the same resolution image. And, I don't derive pleasure by having access to a version of it that other people can see. I derive pleasure to other people getting to see it. I love it and I want other people to see it.
Steve Schindler: Is it the notion of collector as patron in some ways? That you are supporting an artist and to some extent the ownership of it is maybe less important to you?
Jason Bailey: Yeah. There is a shift happening and it's a small shift. I'm not going to say that like the whole world’s moving this way yet, but my life is better when I wake up tomorrow and artists who I like have made new work that I get to see, right. So, if I can buy the work and know that I'm supporting them and you know I own it, and you say, "Well, what is that mean that you own it?" Well, actually, this isn't why I buy, but I can sell it. There is a pretty healthy market. I have outstanding orders for most of my collection that are multiples of what I paid, right. So, there is a market and you can buy and sell through the blockchain, but that's not why I'm buying. I'm buying because I want to support the artists. So, when I get a question, "Jason, are you crazy! Why would you spent 200 dollars on a JPEG that everybody can see for free?" And, it's like well, a) to support the artist and you know b) because it makes me happy and like I do technically own it in that case. So, I don't think — the same way that I still have some books, right and there is a push to like, buy vinyl for people, like this resurgence in buying vinyl, I am not saying that it's an eradication of the physical. I think all those older works that are made that are physical — people are still going to go to museums. Museums are my favorite place. There are still going to be people that paint, just as there are still people that put out albums, just as there are still people that publish books. But, we are talking about a trend, right, not an absolute thing, and I think it's hard to deny that we've seen a massive trend in you know publishing and in music and I just, it seems obvious to me based on some of these early signs that we'll see a massive trend on the art site that follows.
Katie Wilson-Milne: Alright. Well, thank you so much, Jason, for being on the podcast and we look forward to connecting again in the future.
Jason Bailey: Yeah. Thank you so much, Steve and Katie, for having me on. It was a lot fun, and hopefully, I will run into you at one of these conferences coming up soon.
Steve Schindler: I'm sure you will.
Katie Wilson-Milne: Undoubtedly, yes.
Steve Schindler: And that's it for today's podcast. Please subscribe to us on iTunes or wherever you get your podcasts and sent us feedback at firstname.lastname@example.org. and, if you like what you hear, give us a 5-star rating. We are also featuring the original music of chris thompson. And, finally, we want to thank our fabulous producer, Jackie Santos for making us sound so good.
Katie Wilson-Milne: Until next time, I'm Katie Wilson-Milne.
Steve Schindler: And I'm Steve Schindler bringing you the Art Law Podcast, a podcast exploring the places where art insects with and interferes with the law.
Katie Wilson-Milne: And vice-versa. The information provided in this podcast is not intended to be a source of legal advice. You should not consider the information provided to be an invitation for an attorney-client relationship, should not rely on the information as legal advice for any purpose, and should always seek the legal advice of competent council in the relevant jurisdiction.
Music by Chris Thompson. Produced by Jackie Santos.