A lot of the articles I’m reading about AI at the moment seem to make the same assumption: that the future of AI will require vast amounts of centralised compute. Bigger models. Bigger data centres. Bigger chip orders. Bigger energy requirements. Bigger capital commitments.
A lot of product arguments are really arguments about whose reality counts.
The UK does not lack talent. It does not lack good universities. It does not lack creative people, ambitious founders, world-class researchers, skilled designers, clever engineers or globally respected institutions. What it lacks is a coherent economic story.
Benedict Evans has a useful habit of standing slightly away from the noise. For years, his big strategy decks have acted as a kind of weather map for the technology industry: mobile, media, ecommerce, platforms, regulation, capital flows, and now AI. They are not predictions in the cheap sense. They are attempts to show the shape of the system: where the money is going, what assumptions people are making, which comparisons are lazy, and where the industry may be fooling itself.
Most software today is concieved by humans. Even when it contains variations like different checkout flows by country, onboarding paths by customer type, consent screens by regulation, or payment options by market, those variations are anticipated, designed, tested and maintained by people. Adaptive software changes that relationship.
The calls for Keir Starmer to resign are understandable. Labour has taken a serious beating in the local elections. The government looks exhausted less than two years after winning a landslide. Reform is pulling voters away on the right. The Greens are pulling voters away on the left. Starmer’s personal authority is badly damaged, and recent polling suggests the electorate has fragmented in a way that should terrify anybody hoping for a straightforward Labour recovery.
So the obvious response is to change the leader. I think that would be a mistake.