Why left-wing parties are not pushing for GenAI

A manifesto. Ten years ago, almost to the day, Nick Srnicek and Alex Williams, then scholars in their early thirties, published Inventing the Future: Postcapitalism and a World Without Work. The book, coming on the heels of the Brexit vote, reads like a manifesto for human emancipation. It aimed to provide a playbook for contemporary left-wing movements and was met with positive reviews, eventually being translated into five languages. It probably played a role, if only minor, in the rejuvenation of the left in the past few years.

On its cover, the book calls for “full automation” and “universal basic income.” Working less, the authors argue, would free time time for more fulfilling endeavors. By automating tedious tasks, the vision could be realized without lowering living standards. “Tendencies towards automation and the replacement of human labor should be enthusiastically accelerated and targeted as a political project of the left,” they write. Now, doesn’t generative AI seem like the perfect match?

Implementation. I asked Alex Williams, one of the authors and now a lecturer in political science at the University of East Anglia. He told me that left-wing full automation “would require AI and automation technologies to be placed under the control and direction of the people, presumably via the state or a supranational institution.”

This is far less radical than it sounds. In Europe, technologies such as radio or television were controlled by the state from the 1920s through to the late 1980s. Many others still are - such as nuclear energy. If GenAI is as powerful as some of its proponents claim, state control would be hardly out of the ordinary.

Politics. Williams also noted that building such a future would require political parties to address the issue far more forcefully. In the United States, Bernie Sanders is calling for a 4-day workweek funded by a so-called robot tax. In Germany, the Social-Democrats stated explicitly in their 2024 manifesto (p. 38) that “digital productivity gains should lead to a reduction of work time.” However, the phrase disappeared from their 2025 platform. More to the left, Die Linke campaigned in the most recent general election (in March) on the argument that the benefits of AI, automation and robotics should be “shared fairly” (p. 58).

Perhaps not incidentally, this argument was at the very bottom of their 60-page manifesto. Tobias Woelki, an elected official from Die Linke who specializes in digital issues, told me that algorithmic discrimination, management and surveillance were more tangible, and that the party zeroed in on those, when it came to communication. However, over the past two years, the focus has broadened to include issues of ownership and power as well, he added.

Productivity. Before calling for a utopia where human labor is replaced by AI, AI needs to be capable of replacing human labor. While millions have found interesting ways to use GenAI (I even use it to produce a small part of this newsletter!) there is ample evidence that it cannot replace humans on any meaningful scale. Indeed, European statistics shows that productivity grew steadily up to 2020 and declined since.

However, GenAI is a system, and not a machine that could, one day, increase productivity. Both Williams and Die Linke’s Woelki said that a left-wing vision of AI would have to start with realizing that the whole industry was first and foremost a financial enterprise. If the technology ever brings productivity gains, they cannot be detached from the harms it entails. Left-leaning parties could push for public ownership of AI, but the more appropriate comparison might not be nuclear energy, but rather nuclear waste.

This is an excerpt from the Automated Society newsletter, a bi-weekly round up of news in automated decision-making in Europe. Subscribe here.