Introduction

Hello? Ah, okay. Here we go. Wonderful. Thanks a lot for coming, everyone, to our conference, Art in the Age of Average, the New AI Solitarians, which I organized, or Frances Hunger mostly organized, I have to say. And we will do all the thank yous later. Of course, I would like to thank the Academy. I would like to thank all our guests for arriving. I would like to thank our students, who are also co-organizing this conference, and most of all, Frances, who did most of the work. So I'm going to start very quickly doing the introduction, and Frances will do another one, because we became quite ambitious, and really want to do an introduction for once, because it hit us that the topic is actually pretty urgent. And the topic is, of course, the current conjunction of generative AI, or generative art, maybe more broadly. And I I , for the moment, am not really aware of the current forms of authoritarianism. And so we will do two short introductions to just frame the topic, and then I'm looking forward to almost two days of wonderful and very diverse contributions that will try to address this topic, most of all from artistic point of views, which is something that we found extremely, also important to frame this from an artistic perspective.

Let me start with one question. You know, is there anything we could call degenerative art? And I'm not asking this as, you know, by coincidence, because we all know that the term degenerate art was originated by a former colleague of mine who worked at this academy, a guy called Adolf Ziegler, a professor of art who worked here in the 30s, who organized the later infamous exhibition called Degenerate Art in the premises of what is now Munich Kunstverein. And I'm asking myself the question, if there is degenerate art, or if there was degenerate art, is there anything we could term degenerative art today? What could it be? And today, of course, the rise of, as I said, all these cultural movements, all these cultural generative AI forms coincides, or without any doubt correlates, you know, with authoritarian movements. You can think of many well-known examples, such as the recent Trump-Gaza AI slop video, etc., etc. But I think it's way more urgent to think also about the advances in machine learning and constructing huge databases that are supposed to automate government or, you know, to automate government. And I think that's where we're going to have to combine a lot of different other databases in order to, you know, for example, identify and deport migrants in the U.S., for example.

So, of course, if we think of that, then we also can see another development that's taking place at the same time that is the rise of many cultural wars from the right and in order not to overpower them, and in order not to overpower them, and in order not to overpower them, and in order not to overpower them, we always use the same examples from the U.S. This is a recent example from Slovenia, a poster for a referendum from the conservative party over there against some sort of artist pensions. The artist who did this work, Maja Smreka, was basically put on this poster, which says, change for the people, prestige for the elite, to make people, PR for a referendum, which was supposed to deny artist pensions to certain people in Slovenia. The former prime minister of Slovenia, Janas Janča, of the conservative party, has explicitly used the term isrogen, which means degenerate, if you translate it, to refer to non-classical forms of art or this form of art, which he thinks are degenerate, that should not be funded by, by the state anymore. The referendum took place, I think, in May or end of April, and this party won it.

And of course, we have many examples here in Germany as well. Cultural wars from the AFD party against, for example, the Bauhaus in East Germany, and concentration camp memorials, etc., etc. And this conjunction between the rise, on the one hand, of generativeness, of generative art, and authoritarian forms on the other hand, has been described by many authors already. Roland Meyer writes, for example, that AI-generated imagery has become the preferred aesthetic of digital fascism. Other authors speak of techno-fascism. A person called Janis Mimura in the States, interestingly, in conjunction with Japanese imperialism, and techno-feudalism, which is a term coined by Janis Varoufakis, vector fascism, we're going to hear about it in a minute, by Grigory Chautonski. Paul Feiglfeld came up with the term AI-Solitarianism, etc., etc. And all these arguments, they, you know, differ in many places, and they also converge in one decisive point, and this is the point I would like to focus on today, which is that, definitely, whatever it is now, which we don't precisely know yet, it's not the same thing then in the 1930s of the 20th century, because it's just not possible, you cannot jump into the same river twice.

So, in this vein, I think it's important to highlight the new elements of these developments, and not to focus too much on terminology and get stuck in the question of whether this is really fascism or not. I think that's a dead end. This is also not a question for artists to solve. I think this requires historians, frankly. But, in the meantime, from my perspective, I propose some quick and rough and dirty tests, which could also be done by an AI to distinguish different forms of authoritarianism, and this is the so-called duck test. And you all know the duck test. The duck test is, if it walks like a duck, if it talks like a duck, then most probably it is a duck. And I think if we look at this image, we can say that this walks like a duck, and salutes like a duck, I guess, maybe. But, if we start doing this, then immediately we see that it gets a little more complicated, because I, with my old pre-internet brain, I'm immediately, reminded of this old cartoon from the 40s. This is a very famous Disney cartoon showing a Nazi duck actually doing the Hitler salute.

The duck is called Donald, as you can see. And this, if you go back to it now, it's really interesting, because the end is so fascinating and ambivalent. It turns out, you know, Donald is forced to be a Nazi. He's forced to, you know, do all these Hitler salutes, et cetera, et cetera. But then at the end, he's really relieved, because he wakes up in his bed, he's wearing his stars and stripes pyjama, and still he salutes out of reflex, because there's a shadow on the wall. And then finally he finds out that this was the shadow of the Statue of Liberty. And I find this completely fascinating today, because for me it opens up the question, okay, what is the relation? What is the relation between the Statue of Liberty and its shadow? Can we say that, you know, the Hitler salute is kind of hidden already in the shadow of the Statue of Liberty? What is the shadow of liberty? And, you know, this hidden Hitler salute in there, is it a coincidence? Is it a correlation? Is it a causation? And I think all of these questions can be asked when thinking about these topics.

And I think this is why something like the duck test, which is totally non-scientific, okay, is instructive anyway, because it brings up these kind of questions, these kind of ambiguities, and which cannot easily be resolved. And of course, if you go a little bit further in this direction, you realize that if you try to have a duck do a fascist salute by a chat GPT, of course it will refuse to do it. If you spend like one more minute on it, it would of course do it, but there are generators which do it immediately without asking any questions. This was made with the prompt, replicate an image of Elon Musk doing a fascist salute, and flux, which is generator which did this, has a really interesting ideological response to that. So, I mean, this is the first attempt, right, and this is also very interesting. Flux is a German basic product, and it comes up with a sort of GDR logo behind Elon Musk. This is just, you know, because I love making fun of stable diffusion. Stable diffusion has this aesthetic of amputation, and in this case, it can just not decide whether Elon Musk has four fingers or six fingers.

Okay, so, but anyway, I mean, what I would like to do is to go back into history and to compare, again, two images to one another, because I think that there we will go into a more serious and interesting direction. This is, of course, the proto-meme. You all know it. It's a collage by John Hartfield, which from 1932, published a few months before the takeover of power on the cover of Arbeiter Illustrierte Zeitung, which was a leftist magazine published by a guy called Willy Munzenberg, who had a whole publishing empire, publishing leftist and popular populist content, but on many, many different platforms. And as you can see, the so-called salute is interpreted here as receiving a pack of banknotes by someone standing behind Hitler. And of course, this is supposed to be an industrial, like, group, or Siemens or Thyssen or these other guys who paid for the Haus der Kunst, for example. The difference to now is that today there is no one standing behind Elon Musk, right? Because it's just totally unnecessary. There is no one because he has all the money by himself. I mean, he doesn't need anyone. And not only does he have the money, he is also the politician.

He has the public platforms, all of them, no more Willy Munzenberg, no more Willy Munzenberg anymore. And he is the artist because Flux, the video generator you saw before, is now implemented into his platform X, which was formerly called Twitter. And I think that this is one of the main differences between now and today, this absolute unification of functions in just one person, the immense concentration of power, of media power, political power, financial power, in the hands of just very, very few people. And I think that especially this comparison shows us that today we also don't have a kind of system alternative like there was in the 30s. There was a system alternative, which today we don't have. We don't have a counter publishing empire. And we also have no idea what to do after the end of new liberalism. And this brings me to my last point, and I'm going to be very quick, which is we don't have an alternative infrastructure either. We don't have an alternative way to produce generative content in whatever form. We don't have an alternative public platform and so on and so on, whether it's digital or in some cases also the analog ones, as I said in the beginning.

Unfortunately, I have to skip. This is the Italian brain session, which I don't have time for. I don't have time for today, but it's a preview. It will be fascinating. This also contains references to movies in Catalan, et cetera, et cetera. Okay, I'm stopping this because where I want to go is that we don't exactly know is there degenerative art or not, but maybe it's the wrong question. Maybe we have to stop thinking about, you know, the , the thing where the location of the art and where things come from, from which they originate, you know, the genetic lineage of things. Maybe we have to think of these images in a very traditional way as operational images, as images that do things and images that do work and that work on us and that train us and that format us, et cetera, et cetera. This discourse is very old. Where I want to go is that, of course, we can also say this about art nowadays. A lot of art is operative in the sense that it has to be efficient, it has to do something, it has to act on the world, it has to have, you know, deliverables, quantifiable results, blah, blah, blah, and so on.

So my question is, can we just ask another question instead of is there degenerative art? Can we ask the question, is there anything like inoperative art, an art or a production of images which does not work, which is not based on production, which is not even based on the production of community? I'm referring back to a term coined by Jean-Luc Nancy in his Boco collection of essays called inoperative community, where he said that community cannot work. This is my now very short summary. It cannot work if it is made to work, right? It cannot work because otherwise, you know, it will fall back in some imaginary of purity, which requires an enemy, which then needs to be deleted, eliminated, etc., etc. So how can one make in any medium whatsoever images that do not work, that are inoperative? And that's my question for that conference. And now I'm going to hand over to Francis, who will give the second part of the introduction. Thanks for your patience. Yeah, thank you, Hitu, and welcome again to the conference. Yeah, before I start, Hitu and I would like to thank a few people. We thank the Emergent Digital Media class students for helping with the organization and with the catering.

We thank our esteemed colleagues, Boris Shuchkovich and Paul Feigelfeld, who will co-moderate the talks during the conference. We are also indebted to Paul Feigelfeld, who suggested the notion of AI authoritarians, and thank him also for the partner event in Salzburg, which took place yesterday. We'd like to thank the staff and administration of the conference, the Association of the Academy Munich for their infrastructural work, also the media workshop, who borrows us a lot of stuff to make this event possible, and then also especially to Thomas Köhler, who did the sound and media setup. And I would also like to thank our terrific graphic designer, Kathe, who obviously designed the posters and did the graphic design for the conference. And of course, we thank our brilliant lecturers, who took time, who took their time to come and join us here. In my introduction, I'm going to delve into definitions of fascism to set a tone for the conference. And we were debating a little bit whether this was necessary or not, but I think that lots of what I'm going to present here, is well known, but also it might be still a good reminder or an update on a discussion for those who haven't heard that much about it.

So my task in the next 15 minutes is not so much to present anything new, like Hito did, but put forward useful elements of fascism theory as a background for the discussion. So it will be very didactic, but that's what I'm doing. So, in his 1935 text, The Work of Art in the Age of its Technological Reproducibility, Walter Benjamin asserted that fascism tends to aestheticize political discourse. And I'm quoting from his text. He wrote, Fascism attempts to organize the newly arisen proletarian masses while leaving intact the production and property relations which they strive to abolish. Fascism sees its salvation in granting expression to the masses, but on no account granting them rights. And Benjamin also positioned the aestheticization of politics in, and I'm quoting him again, he suggested that there are great ceremonial processions, giant rallies and mass sporting events, and also war. This is where he saw the aestheticization of politics. Similar to Benjamin, we can observe a recent wave of the aestheticization of politics, but now it coldly steals from critical practices, from institutional critique, from cultural critique, and from formerly progressive iconography. Donald Voldemort in his mass rallies, for instance, uses the sign of the fist and appropriates it from the historical worker movement and also from the Black Lives Matter movement.

Both Javier Millet and Elon Musk appeared at public rallies with a chainsaw, a figure appropriated from originally the transgressive popular culture from the 1974 slasher exploitation movie The Texas Chainsaw Massacre. To give another example, Alice Weidel of the RFD floods the zone with shit, that's a strategy by Steve Bannon, with calculated, and provisional provocations like, and I'm quoting her, Hitler was a communist and saw himself as a socialist. And an ongoing myriad of provocations and derailments, as recently in the German parliament where she said, and I'm quoting her again, anyone who upholds the destructive policy of open borders, anyone who upholds the destructive policy of open borders, is an extremist. A clear reversal of terminology. So, formerly critical activists and artistic practices in the tradition of the communication guerrilla of the post-1980s, the adbusters and pranksters that originally were part of an institutional critique are now used by radical right-wingers to dismantle democratic and juridical institutions and to attack elections, to attack court decisions, or attack education institutions. And we are an education institution. Whatever stands in their way of accumulating power is attacked. When artist group Uber Morgan realized their controversial net art piece vote auction in 2000, little did they anticipate that right-wing capitalist Elon Musk in 2025, 25 years later, would publicly give away $1 million checks to voters in Wisconsin in order to buy votes and make democratic elections a joke.

In Russia, we had to observe how society underwent a forced de-democratization since the late 1990s, 2000, and is silenced with fake elections. It has been militarized over the past 20 years, and it's been a very important part of the political process. It has been militarized over the past decades to enable Putin's imperialist oligarchy a radical form of capitalist exploitation. This ultimately led to relabeling Ukraine's democratically elected government as fascists. Once again, a reversal of terminology into its opposite. And I think in this time of reversal or cooptation of formally progressive aesthetics, analysis, and artistic practice, it is worth to recall scientific theories of fascism. During the conference, we will hear analysis such as digital fascism, vector fascism, or AI authoritarianism. And I'm very much looking forward to this discussion, but before I proceed, I'm going to dive a bit more into what can be called fascism theory. The historical periodization of the notion of fascism is rooted in its origins 1919 to 1922 in Italy, and from 1933 to 45 when it spread from Germany all over Europe and parts of Asia. After 1945, the concept was slightly updated by using the term neo-fascism. And there are a number of theoretical approaches which I will now talk about.

So one theoretical approach to fascism theory is the singularistic fascism theory that concentrates on fascism either as a national particularity or the work of one single Führer. Second, we have generic theories that explain fascism through the use of extremist nationalism or as a movement of criminals and the subaltern. Others concentrate on describing fascist praxeologies by analyzing of fascist organization, culture, and practices. Third, there are theories that build on the point of modernization and that refer to social rifts caused by modernization. Fourth, there is a sort of Leninist Marxist theories that include both class analysis and the traditional simplistic common turn reading where capitalists act as agents that try to deceive the working class. A fifth theory is post-war Western theory, especially totalitarianism theories that often, but not always, develop, sorry, not always zoomed in on a similar features of communism and fascism. Developing a horseshoe theory to bolster the Western democratic approach as the only possible countering of fascism. And sixth, there is an undogmatic post-Marxist theories that began to ask the voluntary and tangible of most of Germany's population into constituting a fascist mass movement as the Volksgemeinschaft, combining post-Marxist analysis with psychoanalytical theories.

And I'm going to dive a little bit more into it because I think we will sort of see a lot of elements that we currently can observe. So common to many analyzers is the understanding that fascism is a reaction to larger capitalist crisis. In the 1950s, Max Horkheimer at the Institut für Sozialgeschichte pointed out how capitalist production up to the 1930s had led to significant increases in productivity. And one of the consequences was an overproduction crisis. When not enough profitable sales could be realized anymore in the 1930s. According to Horkheimer, an authoritarian setting enabled those in power to subdue human labor under worse conditions than before, with a totalitarian state stepping in as a violent actor and replacing market-mediated forms of labor with forced labor of foreigners and outright slavery in concentration camps. Friedrich Pollock in 1951 added a psychoanalytical dimension when he described how the fascist German leadership constructed enemies and scapegoats so that the oppressed could sadistically vent their anger. Pollock also addresses the oppressed's desire to subordinate themselves in a masochistic act to fascist domination. So this kind of self-subordination is being read as masochistic. And this resonated a lot for me remembering how at the last conference I described the motives of AI hype as both as sadistic and masochistic othering of machine agency.

Pollock saw the gaining of power as a major motivation in fascism that would outplay the pursuit of profit. Psychoanalytical approaches building on Freud, interpreted fascism as consequences of an authoritarian personality. And here Erich Fromm in 1930 and Theodor Adorno in 1950 addressed the failed process of individuation during childhood. The absence of growing up and forming a self. Authoritarians, they said, feature weak and insecure personalities that are overpowered by their own repressed desires. The authoritarian character feels threatened by changes in society. And I am going to jump a little bit. It was Klaus Teveleit in 1977 who identified some of the authoritarian elements with a specific form of masculinity. A plating and armoring of the self. Forming a male culture of destruction and annihilation. And this resonates a lot with Timnit Gebrou's current recent critical analysis of today's test creel ideologies that is transhumanism, singularity expectations, modern cosmism, rationalist ideology, effective altruism and long-termism. And I think in addition to that today the analysis of the authoritarian character has to be expanded to also include climate crisis denial and the current anti-feminist and anti-queer backlash. So from a political economic perspective I suggest to reread the end of the Cold War and the 1989 breakdown of the socialist countries as a capitalist expansion into the East at that time.

Establishing new markets where commodity overproduction still could sell. Products as a profit. And since around 2010 this market is now saturated and new crisis ensued. So what we experience as AI hype is the necessity for capital to create just another market that returns a higher profit on investments. This AI project is seemingly successful in two ways. What is called AI indeed builds on big data and social platforms. That transform human subjectivity. So they transform human subjectivity into data and later on into value and into more capital. And we can argue that this has established a new market to be conquered as the commons get dispossessed by scraping from the internet what originally was discourse and sociality and now in the hand of the platforms it has become a market. And it's a market that is very much focused on content. That's by the way also a reason why I'm very allergic to the word content because it's exactly the language of this ideology in a way. And the second observation is that populist and proto-fascist actors step right in where capital needs to destroy social rail guards in order to get into a position where social resources can better be exploited.

Today, and I'm coming to an end, in Europe and the US, fascists are aiming for the rain. For them, right now, it is the most unstable moment because interventions still have a chance to break their grip on power. Strategically, it is especially important to intervene, don't ask me how, before fascism constitutes the society's majority. The motivation cannot lie in any hope for success. Instead, it is grounded in the outright and justified fear of the dark potential of machinic agency in the hands of fascists. The application of databases, of drones, of so-called AI against humans like me and you. From an artist's perspective, we need new aesthetic strategies that respond to the situation. And in this sense, I'm very much looking forward to our conference today. And tomorrow. Okay. So this was the didactic part. I'm very happy to now introduce Paul Feigefeld, who will join us together with Hito to moderate the next sessions. Paul is a professor of digitality and cultural mediation at Mozarteum in Salzburg. And he's also co-director of the Open Arts Institute and the Data Arts Forum, also based in Salzburg. And with that, yeah.