In early July we dropped in on the Amsterdam Digital Methods Summer School, and contributed to Gavin Mueller’s data sprint on tankies—hopefully not derailing too much with our AI and climate obsessions!
What’s a tankie? The question kept coming up over the week! The term is a pejorative for a certain kind of authoritarian leftist. It emerged on the British left after the Soviet Union sent tanks to suppress the Hungarian uprising of 1956, and was later applied to those who defended the crushing of the Prague Spring in 1968. “Tankie” therefore overlaps a lot with “Stalinist”—some might call them the same thing.
The terms are not entirely synonymous though, as tankie nowadays has become a looser and broader jibe for leftists excusing repression when it is carried out by an ostensibly socialist or anti-imperialist state … or who are just being so statist about something that you imagine that they might do this. (In other words, if you’re an anarchist or an anti-authoritarian communist, you might have a lower bar for who counts as a tankie than if you are, say, a social democrat).
Tankies might also be understood as a contemporary online subculture, consisting of statist communists who staunchly defend either the historical record of the USSR and figures such as Stalin and Mao, or present-day nominally communist states including China, Cuba, Vietnam and North Korea (AES: “Actually Existing Socialism”). These communities usually associated with an anti-imperialism organised primarily around opposition to the hegemony of the United States and its allies. However, tankie-ism can harden into geopolitical campism—the assumption that states opposed to Western power should therefore be defended almost regardless of their conduct. There is some indication—one of the things Gavin’s research is exploring—that parts of this subculture arrived at being Very Online Leftists by way of being Very Online, at least as much as by being Very Leftist.
Exploring tankie views on AI and climate therefore offers a revealing window onto broader political struggles over technology, ecological crisis, empire and competing visions of modernity.
Gavin and Ema had prepared two text corpora, scraped and transcribed from YouTube videos—one from independent YouTube tankie influencers (‘YouTubers’) and the other from traditional Marxist-Leninist organizations with a YouTube presence (‘organizations’).
At first we used a randomly downsized data sample, too small to draw firm conclusions. It looked plausible that ‘tankie’ YouTubers think about AI and climate a little differently from the wider online left.
This corpus didn’t come out as either consistently pro- or anti-AI. YouTubers celebrate AI’s potential to reduce toil and raise productivity, while criticising environmental costs, job displacement, and dehumanising effects. More distinctively, they sometimes read AI as symptomatic of the contradiction between highly developed forces of production, and increasingly obsolete capitalist relations of production. Some use AI tools artistically—for example, to depict Marx as a bodybuilder.
Most distinctively of all, AI is also frequently linked to geopolitics—US, China, Russia, Ukraine, Israel, Palestine. This includes the IDF’s use of military AI in Palestine, and AI’s role in dis/misinformation and genocide denial. Tankies may contrast US and Chinese approaches, presenting socialist AI as more economically competitive and socially beneficial, sometimes linking this to China’s open-source model.
A crude estimate of how much AI appeared in a ‘geopolitics’ frame (how often do any of a set of AI-related words co-occur with any of a set of geopolitics-related words?) gave a fairly striking result: organisations 19% of the time, YouTubers 35% of the time.
Nathalia Henao beautifully visualised this, along with other findings, for a poster presentation:
Next we went back to the big dataset, and downsampled in a more targeted way to create four datasets: individual YouTubers talking about climate, individual YouTubers talking about AI, organisations talking about climate, organisations talking about AI
Then we tried some topic modelling. You can find the details here.
The short version is: there were no climate-related topics in the AI corpora, and no AI-related topics in the climate corpora. Does this contradict the earlier finding, where AI was sometimes criticized for its environmental impacts? Not exactly, it just suggests that in these different and slightly larger corpora, climate and AI things are not correlated frequently enough to form a distinct cluster.
In the broader social resistance to AI, environmental impacts loom large. Perhaps this is partly to do with carbon pollution and water consumption putting a clear cost on AI, which might act as a proxy for more nebulous unease, where a popular vocabulary is lacking — enclosure of knowledge commons, conversion of common social capacities into proprietary models, growing dependence of private and public life on computational systems controlled by a small number of corporations?
Similarly, it may also be to do with the readiness of the environmental movement to incorporate critiques of big tech into existing messaging, and the responsiveness of the environmental movement as a concrete set of actors, with knowledge of a specific repertoire of contention–campaigns, protests, people’s assemblies.
According to this broad approach, the problem with AI is that it is being developed and deployed by a profit-driven, growth-oriented and extractive industry with extraordinary influence over the institutions ostensibly responsible for regulating it. For some, “regulatory capture” describes part of this relationship, but doesn’t go far enough: critiques are levelled at the shared worldview of governments and big tech.
This is also a very situated view. From our position as researchers based in Europe on the BRAID Sustainable AI Futures project, the environmentalist framings most visible to us are often accompanied by degrowth or post-growth politics, demands for democratic control over infrastructure, and versions of libertarian municipalism associated with Murray Bookchin, with perhaps some decolonial influences and the valorisation of Indigenous perspectives. These tendencies are important, but they should not be mistaken for a clear and comprehensive map of resistance to AI.
The equivocal term “system change,” associated with the phrase “system change not climate change,” nowadays can name anything from modest regulatory reform to the abolition of capitalism. Equivocation is not necessarily intrinsically bad–alliances can be formed around ambiguities, and sometimes it is important to keep things open, especially when the opportunities for real change look thin on the ground. However, alliances and coalitions based on ambiguity can also be especially frail, and once grassroots resistance to AI begins to specify which system must change and how, previously submerged conflicts may become unavoidable.
The picture is complicated and messy. Opposition to data centers in the USA also includes organisations like HumansFirst, whose chairwoman was a co-founder of Tea Party organizations and a pro-Trump PAC. As for the tankie communities we glimpsed here, they may be more likely to see “alternative AI imaginaries” as something that already exists in practice, contrasting Chinese approaches with US/Europen approaches. They may also be more likely to consider AI harms in terms of geopolitical rivalry between capitalist imperialism and socialist counter-imperalism, and contradictions between advanced productive capacities and obsolete relations of production … rather than as reasons to reject AI, or seek changes in its ownership and governance.
Special thanks to the Sussex Digital Methods Accelerator and HEIF for supporting participation in the Summer School.
