Arts Of The Working Class Logo


A discussion with Hito Steyerl and Mat Dryhurst, moderated by Joshua Citarella, on November 21, 2022 at the Akademie der Künste, Berlin.

  • Mar 20 2023
  • Hito Steyerl, Mat Dryhurst and Joshua Citarella
    Hito Steyerl is a German filmmaker, moving image artist, writer, and innovator of the essay documentary.

    Mat Dryhurst is an artist and musician. He lectures on issues of music, technology and ideology at NYU, and advises the blockchain based platform co-operative

    Joshua Citarella is an artist researching online political cultures.

Joshua Citarella: Where is the public? In the last few years, we have seen tech platforms radically reshape creative economies, legacy media, cultural institutions and society at large. Today, when we think of “the public,” we unavoidably and necessarily think of data. But does this data, as captured by our privately owned platforms, represent any real sense of society? Is there even a public without a commons? Do new forms of collective ownership offer solutions to the numerous crises facing us in the form of skyrocketing rents and education, austerity, informational chaos, and social disintegration? What role might institutions play in shaping this new society? 

I was told by the artist Rachel Rossin that “the metaverse,” as a term first appears in Snow Crash (1992), the dystopian sci-fi novel about an anarcho-capitalist future. One might imagine that there is very little “public” in the world forecasted by Snow Crash. Our panel, our program, began in Bonn with a keynote address by New Models’ Carly Busta and Lil Internet, in which they describe the implosion of legacy media and the rise of platforms. 

I wonder if we might start with a bit of background from each of your respective fields. How has online distribution shifted creative practice? When did you have the feeling that something fundamentally changed? 

Mat Dryhurst: The feeling that something had fundamentally changed was toward the end of the aughts; kind of like the birth of optimism. There was a lot of excitement around the experimental record being presented on the same stage as Beyoncé. Of course, that then soured when we realized that all such works are judged on the same plane; that everyone must become Beyoncé. That was the double-edged sword. With that kind of populism, you have, ultimately, this expectation where things are being judged by how much they're being engaged with. You can describe this as a kind of philistinism, because ultimately, the things that are the most attention-grabbing are the things that get the most attention. My concern with that is that, institutionally speaking, the same tools that push that economic logic started to challenge institutions like journalistic publications. In the past decade or so, there’s been a slide toward attempting to appease a public that may or may not actually exist, at least in music. The one safe space from all that was the live music industry, because it was determined by physics, in a way.; it was sustained by ticket sales. Still, slowly, we saw music from somewhere being displaced from by music from nowhere: bands that thrive in platform environments, like large festivals.

Citarella: Hito, how has online distribution shifted the field of art? 

Hito Steyerl: There's been multiple ways over the last 20 to 30 years, but I’ll focus on the present. What we're seeing now, just in the past weeks, is the unraveling of many of the paradigms we took for granted: namely, some of the major platforms of platform capitalism are imploding at breathtaking speed. I think this will have major consequences for everything we're talking about. I'm calling this The CCK Collapse. C for crypto, C For corporate media — unfortunately, until now, only Western corporate media; not TikTok — and K for Kanye West. All three of them imploded, basically, in a flurry of completely bizarre plot twists. I think only one of them is going to come back, and it will not be Kanye West. 

In the last weeks, we have seen a major implosion, again, of the crypto market by way of the bankruptcy of the crypto exchange called FTX by its CEO Sam Bankman-Fried, who also was a follower of a very weird, pseudo-philosophical, neocolonial cult called Effective Altruism. This is not the beginning of the crypto collapse. It started happening earlier this year: all sorts of crypto institutions, if you can call them that, have already imploded. This is another major blow to crypto’s rhetoric of decentralization, like the removal of gatekeepers to provide opportunities and income for underserved people. 

And the second is, of course, corporate media. I think everyone has noticed what’s happening at Twitter: it's either falling apart or it's really coming into its own, to be a sort of optimized platform for monetized toxicity. This is also flanked by a much larger set of layoffs across the tech sector: Meta’s metaverse is not going to happen any time soon; there have been more than 11,000 workers fired. We have layoffs at Amazon; at machine learning companies. That doesn’t mean that Western corporate social media will disappear any time soon. They will still exist as silos. Facebook has been a silo for frustrated Boomers for a while, and it will continue to be so. I think these silos will just shrink; sort of trickle and wither away, as Marx said a century ago about the state. Or they’ll dissipate into even smaller silos, like Telegram. 

This means that the landscape of criticism that we have heard a lot about during the last days is — very, very interestingly — dramatically shifting once again, because many of the material and informational conditions are shifting as we speak. This is not only intrinsic to tech: this has a lot to do with the ceasing, basically, of quantitative easing; with the diminishing of excess liquidity, which could be invested in crypto or NFTs or art for that matter. And this happens in the context of war, pandemia, energy riots, crypto bust. I think this, for art and art criticism is not just bad news. We’ll see how this will work out. 

Many of the last year's big art debates were characterized by something that I call “toxic flow.” Toxic flow is a term from crypto: it’s when two bots trade against one another on a crypto exchange by using high-frequency trading algorithms. They try to destroy one another. Toxic flow is the dynamic that both legacy and corporate social media developed with one another in trying to maximize outrage and compress facts; they were maximizing polarization and egging one another on in the process. If there's less toxic flow, that's not bad news, honestly. 

On the other hand, there's also massive fallout for art workers who have invested a lot of their time, attention, labor, and energy into building a voice, some influence, and income through these same platforms. For them, I think this must be a major rug pull. It might be really tragic. On the other hand, there is no way to dissuade people from investing their labor and attention into corporate social media, because there's hardly any other realistic option. If there is no decent paying job, then you are not going to berate someone for buying a lottery ticket. That's basically where many art workers in precarious positions begin investing their own time and lives into social media. 

Still,  people are leaving social media in droves right now. They arrive on Mastodon like survivors of a plane crash, completely mentally disheveled, in a vile mood, agitated, yelling: “Where are my followers? Why is the UI so shitty?”  

I see that a lot of the energy of the debates is now shifting into the so-called “fediverse,” which is ambivalent. The fediverse has many of the problems that corporate social media have as well — representation, remuneration, content moderation and so on. But for the first time since Web1, it’s happening in an area which is nonprofit, or not primarily for profit. That's really good news - it shows that people are growing tired of the corporate web and its tsunami of toxicity and that's something I think that institutions should support. 

Dryhurst:  With all respect, I disagree with so much of what you just said. One of my frustrations, listening to the previous discussion, is: at what point do we, as a community, intervene? At what stage in production do we intervene, in terms of critique? 

It's abstract to me, even, to be on Twitter right now, seeing people suddenly embracing the fediverse. People experimented with the fediverse in 2010, and it was a disaster for numerous reasons; privacy being one of them. For example, do not join a Mastodon server with anybody that you wouldn’t want reading your DMs. A lot of the original builders behind crypto — which by the way, the idea of the FDX proves anything but the original principles of crypto are incorrect. FDX, as with Terra, as with Luna, are centralized organizations. The Ethereum network is running just fine, Uniswap is running just fine. Actually, many of the principles of privacy baked into these early crypto networks are there because of the failures of the fediverse. A lot of those original developers eventually migrated into the Ethereum ecosystem, for example, for that reason. We can have a conversation about whether or not Twitter will go away, but the idea that we should move off into the utopian space of the fediverse is not good advice.

Steyerl: But that's not what I said. I pointed out all these problems, and they are many. Mastodon is not privatizing your content. 

Dryhurst: Well, as a result, it will also fail. 

Steyerl: We'll see about that. But it is not, for now, a corporate silo where whatever you do there is already being expropriated by the platform. Mastodon creates some kind of decentralised feudal situation which is already better than to be expropriated by default by some imperial monopolist.

Dryhurst: Until recently, Mastodon ran on less than $1,000,000 annually. The idea that that will compete with whatever Twitter has planned, or whatever the Ethereum network puts forward, is not realized. 

Steyerl: I know you disagree. It's fine. But let's not get into the toxic flow; these debates are set up for maximum polarization and I don't want to replicate this dynamic. But we still need to talk about the failure of crypto to deliver on its multiple promises. I was very cautious with my mention of Mastodon, because the problems are really obvious. But I think it's also striving to create a public sphere for the first time in almost 30 years; one that is not entirely commodified and privatized from the get-go. And maybe scale is not even the point. Why? What for? For whom? 

Dryhurst: My bigger question is, let's say Twitter does die, right? Let's say crypto does die eventually. But then, what replaces it? And as you said, and to give you a lot of credit, you're then saying, look, people buy lottery tickets, right? There is a general sense of precarity. But then, what do we offer people that's actually realistic? From a pragmatic position, I have to defend some of these networks and not so easily dismiss them, because I don't see anything else coming to replace them, if that makes sense. 

Steyerl: I read this interview with one of the people who got scammed by FDX, who said, “Oh, that's just crypto. I don't have a job. I'm not going to be able to buy a house. What else do you want me to do?” That's fairly rational. But because there is no other alternative, we remain hostage to the silo of the expropriation of labor, attention, data, resources, et cetera, by platforms, including crypto platforms. I think that looking for a nonprofit alternative is a good idea. 

Dryhurst: I would agree. I think that the nonprofit part may present challenges in terms of actual viability, but we completely agree on fair remuneration and fair acknowledgment of labor. I just feel like crypto is something of an easy target.  

Steyerl: Well, it's easy to target because it's such a sitting duck. Just think about what effective altruism is; I mean, it is ridiculous. It's a sort of neocolonial, philosophical front for ponzi schemes.

Dryhurst: I'm not advocating for effective altruism. What I’m saying is, I think a lot of early crypto people joined the space from places like the fediverse because they were concerned about the same things that you are currently concerned with.

Citarella: In my observation, from extensive interviews with young people — who are very politicized all over the spectrum, especially young emerging artists — people seem to be opting out of these giant platforms like Twitter and Instagram. They're spending more quality time in spaces that are private but have very low barriers to entry. I think there's a hazard of over-indexing how impactful or meaningful a platform like Twitter is, which is overwhelmingly dominated by journalists, and 10 percent of its users produce 90 percent of its content. It's massively asymmetrical. I don't want to forecast the end of Twitter, but there's certainly a lot of young people who find it not worth their time, and they are choosing to opt in to other networks. That allows for all sorts of conversations about what is an equitable distribution of resources in these spaces, whether they are nonprofit or for-profit, if there is a way to have an ownership or a stake in that community; things that, if the Web3 side ever does happen, we can get back from Web1, because I think, broadly, we are all disappointed with the way things are today. 

Citarella: Let’s talk about one slide that New Models shared, which plots the devaluation of rate per word for journalists between 1970 and 2010. In the 1970s, journalists would get roughly $10 (around €9.30) a word, and in 2020, you're getting $0.25 (€0.23) a word. The title of our conference is The Future of Critique. This framing implies that institutions are anxious that they may be losing influence, and that the future of critique is somehow uncertain. Who are the stakeholders in this space? Who are the institutions losing influence to? Have they been, to some degree, complicit in this hollowing-out of the public sphere? 

Steyerl: It seems as if we're taking for granted that there is a future of critique, there is a future of art, there is a future of the internet.  The new internet: maybe no internet. We are taking all of this for granted. I'm collaborating with a colleague in Kiev, and the internet can be sketchy. This is the new normal. 

As we speak, the Turkish air force is bombing Rojava because they adopted the playbook of the energy wars: taking out infrastructure, heating, and communication. I was researching with three colleagues about crypto-colonialism: about the way crypto mining, for example, is implicated in power networks, like energy networks from Kosovo to Kazakhstan to North Korea. Crypto is basically mined on the back of political instability. It became really clear that in talking about anything related to the internet, digital, data, or even crypto, it's just a subchapter of a more general question of Extractivism: extractivism of energy, data, force, fossil fuels, labor, attention. This needs to be factored into the question, what can institutions do? How can they engage with this extractivism, which plays out on many levels? It’s clear that we cannot go back to this legacy criticism model, because it clearly has been elitist and exclusionary, and, especially in Germany, has never properly dealt with its historical foundations. Werner Haftmann, the SA member, presumed war criminal, torturer of partisans and mastermind of the first three documentas is a prime example when it comes to that. Did legacy art criticism bother to engage with his career and to figure out the implications for Haftmanns influential ideas about modernist art history? Nope. Maybe legacy art criticism should reconsider its pretense to excellence, mastery of context and attention to detail. The blind spots are so glaring that you wonder what kind of vision is left, if any.

Citarella: I agree that it’s important to critique the history of legacy institutions and how they have devalued creative production in music, journalism, video, and perhaps art. Defending the importance of institutions now is not meant to valorize all of their historical baggage, but to forecast how dystopian culture will truly be when the institutions themselves are eliminated, which would mean being left to navigate the pure liquidity ancap dystopia of platforms to find meaningful culture. 

My contention is that we stand a better chance as people with left political dedications to wage those conversations; those struggles within the framework of institutions, than we do within the platform, specifically because of their extractive nature. Do you have a vision of what art institutions are supposed to do in this context? If platforms are understood as the institution’s largest competing force, that framework supposes that when the institutions wither, the platforms will immediately move in and take their place. 

Dryhurst: My concern fundamentally with critiquing institutions is relevancy. I think everyone's in a very precarious, unusual position where, on the one hand, you want to appease a public whose attention is elsewhere. My frustration with many institutions is that we're not intervening early enough in the process. I believe that institutions should begin intervening earlier in the process, when things are actually being built. That could bring a new era of relevancy for institutions. 

Citarella: Are you suggesting that institutions should be diagrammatic and lead as designers, or that they should have some type of input on the platforms for cultural distribution? 

Dryhurst: In the crypto or the machine learning fields, these were public debates for many, many years. Nobody really cared until NFTs came along. Yesterday, I was on a Discord with hundreds of thousands of people debating [text-to-image AI generator] Stable Diffusion’s policies on how they're going to train their next model. So, these discussions are happening; it's just that, in art institutions, nobody cares until there's something to critique years later. 

Steyerl: I would love to have a discussion about Stable Diffusion, but maybe at a little later point. I do have two comments toward these questions of the institution: one is more productive, and the other maybe less. Let's start with the more productive one. 

I was talking about everything being a function of extractivism. Institutions are partly able to reverse this kind of extractivism, as long as it concerns their own internal economy. What do I mean by that? My colleague Paul Feigelfeld had a great idea: that one needs to retrieve the energy stuck in NFTs, memecoins, and Bitcoin. Once it is unstuck one can solve the energy crisis with redirecting the energy that's wasted and frozen within these so-called assets. I would love to transfer this power to Kobani and Kharkiv and Kiev; places where this energy would really be needed. But it's not physically possible. Still, you can take the structure of this model to try to reverse extractivism within an institution. Institutions, our institutions, take for granted the unpaid labor and attention and contribution of many of their constituencies, which are mainly artists, art workers, content providers, mediators, and critics. But these constituencies contribute a lot to digital corporate platforms for free. They contribute a lot to the artwashing and reputation-washing of dubious sponsors, and they do so for free. I think one can really reverse the flow, investing and infusing resources into people who have been extracted from in the past. Infusing resources into these constituencies would take away from corporate platforms by not providing free content or reputation-washing for oligarchs and sponsors. Why should public institutions pay social media managers but not so-called content providers? Why not just walk out from paying to deliver content to platforms for free? 

I think that's one very, very reasonable and pragmatic suggestion. 

And then, there’s the other issue: during one of the talks in this forum, Robert Kudielka pointed to a statue in the corner. I realized that this statue represents Friedrich Wilhelm the so-called “great electoral prince” of Brandenburg, who founded this academy, and also the art school and many other things. It has been public knowledge for many decades that he also traded and sold around 20.000 enslaved people across the Atlantic in the 17th century through his Brandenburg-African slave trade start up. So, we're looking at a slave trader guy here, and it’s completely normalized. Let's not even start talking about serfdom, which in many places ended even later than outright slavery. The “Kurfürst” is part of the furniture that is patronizingly, benevolently, looking upon all of these discussions about art critique and criticism, upon polite debates about their decolonisation etc. And no one, apparently, questions this patronage, or rather the material infrastructure framing these debates, neither legacy art criticism nor the shitposter commentariat.  As long as institutions are not able to deal with this sort of very, very blatant juxtapositions, this cognitive disconnect I'm not sure how much of a positive impact they’ll have on other aspects of art criticism. Lets connect the dots here and question ownership and property relations.

Citarella: I'm from the States, and the rates for young art critics and artists are extremely low, often unpaid. In 2008, the middle of my college career, it was very normal to do two years of an unpaid internship, assume a quarter million dollars of debt to get your education, and then go into a job that maybe paid you $35,000 a year. What you find is that, when you set up those extreme financial barriers to entry, the only people who can afford to do these jobs are those with access to intergenerational wealth. And then we wonder why there's a lack of representation in the art world, because only the wealthy can afford to do it. 

Mat, you brought up this question of the relevance of institutions, and I want to throw in a point of contention. What I often hear from my friends on the stodgy Orthodox Marxist left is that institutions move slowly because we would hate to see if, for example, a museum gallery was devoted to the Baby Yoda meme. There are things that are not relevant and can be better left to magazines and other forms of cultural commentary. There's an important sorting the signal-to-noise ratio that has to happen for meaningful programming to occur in museums. Is there a benefit to the slow pace of institutions, in some instances? 

Dryhurst: The question that always comes to mind is, is critique just something we do? It’s not just something that happens in its own kind of industry, right? You go to school for it and then you continue doing it. We still have these 20th century industrial habits and 20th century educational habits that haven't really modulated over time. For me, it's mostly a latency issue and a fluency issue. Nobody taught me this stuff in art school. It's stuff that you can only learn. My concern with the licenses that I'm seeing in this AI conversation is that, if you miss the tech development for three weeks, you're catching up because something has already changed. That's the vacuum there, which is not to be confused with a desire to start seeing institutions on TikTok. It's not a populist play; it's more saying that we have a latency issue. 

Steyerl: I agree that diffusion networks will be really important going forward, not least because they will automate many jobs and become default apps inside standard software for image generation. Sometimes institutions do not wait though if they smell money, and they can be really quick. In the middle of this full on crypto crash, MoMA is having a huge NFT show, by someone who is probably the last artist represented by Johann König, where they basically create a secondary market for their own proprietary data to be repackaged as NFTs. MoMA smelt liquidity and they went all in in a blink.

Citarella: Yeah, we discussed a little bit earlier that there are such low fees, particularly for young people entering into this space. Let's say a talented young person who's 19 or 20 is looking down the barrel of an art career. We're now a generation older than them, and they see that we are not living very luxurious lives. They think, “I don't know anybody who's made it in this field. Maybe I'll try YouTube. Maybe I'll try something else where, even though it's grinding, you can at least meet your overhead.” You have a chance for success.

On top of that, we've also seen a shift in institutional priorities, where they've become increasingly reliant on private donors as state funding dries up. Although museums are ostensibly  making exhibitions for the public, they are increasingly beholden to the interests, whims, and tastes of collectors, which presents a difficult terrain to try and navigate. Institutions are finding out that it’s no longer true that value is created by placing an artwork in their context, a value which is then later realized through the private market. This is the concerning thing. Because even from the most impoverished position, to say that criticism or institutional prestige is a value add to the rising asset price of an artwork — and as an art evangelist, I hope it does a whole bunch of other things, but the bare minimum definition is that it does that. This is forecasting some type of Snow Crash complete market liquidity, in which institutions are now considered a distortion of price signals, this is the libertarian rhetoric underpinning a lot of these big tech platforms now working their way into the institution, and, through the coercive laws of competition, as art is driven towards an asset price, we're finding that the influence of institutions is really not that impactful. 

I hope that we can address the experience of young people entering these fields, into what seems to be decaying at a rate so rapid that we can't even do the panel discussions as fast as the damage is happening. It’s kind of what Mat said, that some of these decisions, especially in the field of AI, operate at a pace that can be incredibly impactful, for better or worse, on the creative paths of a whole generation. What impact can institutions, can art discourse, have if we're increasingly irrelevant? How do you hit the brakes on this thing? How do you grab hold of it? 

Steyerl: That's nice. Being irrelevant. Sounds good. No toxic flow!  

Dryhurst:  I spent a lot of time teaching older teenagers. You never want to be that kind of cliché of the professor saying, “Oh, well, you know, only a few of you are going to make it.” Of course not. There's a real challenge in that a lot of the narratives about how the world works, about how creative economies work, haven't really been updated. It becomes more and more difficult to talk to younger people with a straight face saying, “Well, this is the right way to do things, even though there might not actually be a career in it for you — if that's what you're looking for.” Of course, I don't think art school was ever designed to necessarily facilitate a career for people, but it's increasingly becoming part of the conversation. I end up having to be an unfashionable kind of optimist in saying that there are plenty of ways to intervene in things that are happening currently. And it might well mean that a future art market will look quite different to the one you’ve read about in books. 

Citarella: Is there a crash in the art market? We’re seeing that there are crashes in other markets and then the resulting polarization is often quite good for art, because art is a way to store your value in anticipation of this collapse. Also, when the internet goes down, museums are quite resilient, because you have things like the institution’s physical structure and discursive networks. 

Steyerl: I think the art market might now finally start to feel the crisis that's been basically ongoing since 2008, from which it’s been, miraculously, incubated for all this time. It sounds crazy to talk about interest rates in this kind of discussion, but I think it will really affect the art market if interest rates rise. There is less excess liquidity to be invested in these so-called “luxury goods.” So, I think, yes, if you look at the data it should tank, but on the other hand the market is so relatively small that a few whimsical buying sprees of not very many ultrarich people make quite a difference. I am very confident that polarization within the market will continue though - benefitting the same few bluechip painters at mega-galleries and successively cutting out mid-size and small galleries. 

Citarella: Mat, I'm wondering if you can tell us about spawning, the AI tool you co-created with Holly Herndon. Can you sketch out what would be the desirable traits for this kind of design framework, considering what we’ve been talking about? 

Dryhurst: For about ten years, my obsession has been the ability to set permissions on your work. With the crypto moment, we had more robust tools to be able to do that, for better or worse. In the context of machine learning, this topic has become relevant because you're now finding that pretty much all artists are, in some ways, contributing to systems that they have no say in. It's quite difficult to actually unpack quite how much living artists’ work contributes to final outputs, but let's just say that it contributes something significant. The goal of the organization is really to build tools that give artists agency over their work in those ecosystems. 

What that means in practice is that, in the short term, we've built tools to opt out your work. We've received commitments from Stability and LAION and a few others coming to honor those requests. The longer term ambition is to think more deeply about the idea of owning your identity in whatever this ecosystem is. There are terms that might not be familiar to many people here, but I think are worth looking at things like embeddings; the idea that you could have images spawned in your likeness or from the kind of work that you do. On principle, it's fundamentally important that you have as much agency over that work as you do in physical life. I'm certainly not a metaverse freak; I've rolled my eyes about a lot of it. But I do think that there are enough signals to suggest that it’s going to be really important to build for actually taking ownership of your data and understanding that that data will actually play a consequential role as a part of you in these ecosystems. But as I said, the fortunate thing is that, in putting forward that narrative and saying, “Look, on the base level, we need so many consenting systems,” so far, the response has actually been pretty good. 

Steyerl: I think in all of these debates, there’s the dual question of retaining some control over your own intellectual property, and letting go of it; letting it be part of the commons, or a common historical repository. I think this balance is going to be crucial. If fellow artists want to appropriate parts of my own IP, why not. It happens all the time and it's usually entirely productive. I would gladly sign over rights to derivatives to an artist commons or data trust if it existed and use proceedings for p2p art funding - but I am opposed to Meta surreptitiously hijacking and privatizing it. We need some exemptions from IP - the case of photos of looted archeological artifacts is an important example. Should anyone respect copyright claims of any Western museum over photos of stolen material from Mesopotomia? I don't think so. Historical or educational material? Needs to be exempt from IP claims by default.

Dryhurst: I couldn't agree more. That's one of the things that I’m most fascinated about in this space: in essence, it challenges a lot of these cozy ideological positions that we developed over the 20th century. You don't want to live in a world where you go to an exhibition and no cameras are allowed, and, at the end, the institution makes sure that you don't retain an image of it because any piece of media can be used to train a derivative work. And that is now the reality. 

Citarella: What types of equity and remuneration would you like to see go into new institutional structures? 

Dryhurst: I've seen a couple of things that I feel quite good about. Recently, I've heard of galleries starting to say, “What if we were to take some of those principles and apply that to a stable of artists where we could do novel economic arrangements; where there was a royalty built in; where all sales would contribute to the group in some way? Let’s make these new economic systems somewhat plastic. Whether any of these work, we'll see over time.” It seems to me that a lot more people are entertaining these ideas of building bespoke economic systems for what they're doing. 

Steyerl: Whatever is developed now also needs to be viable offline, or with a minimum viable configuration of technology. Otherwise, we will prepare for something that either is not in our hands because it works at a gigantic corporate scale, or for a situation which is going to be very different from the trajectory that we came to expect from the 2010s, essentially. Uneven power, uneven energy supply, uneven internet. It's wise to prepare for minimum viable tech - the guerilla version of corporate tech and cloud leviathans. Plus, we need to retrain our own goals and expectations.  We are talking about training data and training ML models, but in essence it's us that are being trained to take this whole energy wasting environment including its radical class hierarchies for granted. Maybe we need to retrain ourselves using inverse reinforcement learning to just leave behind the idea of optimizing for mean tech. Let's retrain to run artist data trusts instead.

Dryhurst: I agree.


    Image generated by the AI text-to-image program Stable Diffusion, based on the text prompt A sword made out of a gummy bear hovers above a sad looking child who is dressed in a sunflower costume. Landscape perspective, 2023.



To improve our website for you, please allow a cookie from Google Analytics to be set.

Basic cookies that are necessary for the correct function of the website are always set.

The cookie settings can be changed at any time on the Date Privacy page.