Thanks to POLITICO Pro Morning Technology UK for asking me to respond their weekly Q&A for the Friday 13th(!) edition. As they put it in the newsletter intro, "AI academic Dan McQuillan has some… strong feelings about the U.K.’s AI strategy".

Q&A

Dan McQuillan, senior lecturer in critical AI at Goldsmiths, University of London and the author of “Resisting AI: An Anti-fascist Approach to Artificial Intelligence,” gave MTUK a radical take on the U.K.’s AI strategy, saying the government is over-committing on the technology.

What are you working on at the moment?

I’m working on getting people to question the AI-for-everything fever dream that has overtaken organisations and government. Transformer models don’t really do what’s claimed of them, but they do undermine jobs, the environment and our capacity for independent thought. I don’t believe most people in tech want to erode the institutions that have made the U.K. a decent place to live, but going all-in on AI is going to help finish off the NHS, higher education and the welfare state.

I’m developing an alternative approach that I call ‘decomputing’, which doesn’t mean getting rid of all computers but does mean doing without carbon-emitting, hyperscale data centres, and it does mean avoiding AI for anything that has social consequences.

Are you feeling optimistic about the U.K. tech industry?

I’m concerned that the U.K. tech industry is going to end up on the wrong side of history. We should be worried because AI can’t fix our important services or make most people’s lives better, but we should be even more worried about the backlash. When struggling people are fed a vision of the future that flops, history tells us that the far right are waiting in the wings, and there’s enough of that kind of politics in Silicon Valley already. We can still imagine tech for the common good, but it looks like that’s going to have to come from the bottom-up.

Where do you think the U.K. government is doing well on tech policy? Where could it be doing better?

The U.K. government is probably the most AI-pilled in the world. Every time I think I’ve heard it all, they come up with another humdinger. The recent AI Action Plan was like a second-rate startup pitch, even if you ignore the possible conflicts of interest behind it. Meanwhile they’re willing to stuff the regions with data centres and dump on copyright, just because Big Tech told them to. They’d do better by focusing on low tech fixes first, like people in our public services are asking for, rather than handing our data over to the likes of Palantir.

What U.K. tech policy/legislation are you most focused on at the moment?

I’m afraid that current legislative proposals rather prove my point. The combination of the Data (Use and Access) Bill and the Fraud, Error and Recovery Bill are a literal recipe for repeating Australia’s ‘robodebt’ disaster at scale. Honestly, it’s like we’ve learned nothing from the Horizon IT scandal. But what I’m really wary of is the signs of DOGE-like thinking creeping into U.K. circles. Now that Musk’s mask is off, we might hope that people would be more wary of muscular tech ‘solutions’ to people’s complex needs. What we need is less sci-fi hype and more care for each other.