A leaked Pentagon memo suggesting the US might oppose Britain’s claim to the Falklands. That’s the kind of blunt force diplomacy Donald Trump respects, a far cry from the soft power King Charles’s D.C. visit aims to project. But this geopolitical saber-rattling is a smokescreen for a deeper, more existential threat to British sovereignty: its burgeoning dependence on American artificial intelligence.
Here’s the thing: the current White House operates less like an ally and more like a protection racket. The price of admission? Sovereignty. Give them military bases, tax breaks for their cronies, even try to buy Greenland. Say no, and face the consequences – like potential trade deal binning and new tariffs.
This mercenary approach isn’t new for Britain. The US has always been the domineering sibling. But the stakes are now astronomically higher, especially as the US, in its intensifying rivalry with China, is rapidly pulling away from Europe in terms of technological might. The currency of future power, as science secretary Liz Kendall put it, is AI. And Britain risks becoming a vassal state, beholden to a handful of American companies that control the vital digital infrastructure.
This isn’t just some abstract future worry. The latest iteration of Anthropic’s Claude model, Mythos, is so adept at finding code vulnerabilities that its creator has restricted access, acknowledging its potential as a cyber weapon. While some cynics might dismiss this as marketing theatrics or a simple lack of computing power, the independent verification of Mythos’s hacking prowess is undeniable.
Even if you correct for marketing hype and take grandiose forecasts of godlike intelligence out of the equation, the machines are getting fiercely good at tasks that frighten politicians.
And let’s talk about the architects of this future. Dario Amodei of Anthropic, while seemingly more safety-conscious than some peers (his company was blacklisted by the Trump administration for refusing to license its AI for autonomous lethal weapons and domestic surveillance), is still part of an ecosystem where ethics are, at best, a secondary concern. Sam Altman of OpenAI, the architect of ChatGPT, reportedly embodies a ruthless ambition bordering on the sociopathic. Then there’s Alex Karp of Palantir, whose company’s manifesto explicitly rejects ‘vacant and hollow pluralism’ in politics, making it clear their mission is to serve US economic and cultural supremacy – a mission broadcast on X, Elon Musk’s platform, formerly Twitter, a place where Musk’s own AI chatbot, Grok, once identified itself as ‘MechaHitler’. The mind reels.
Why Does This Matter for UK AI Development?
Liz Kendall’s speech, while likely overshadowed in the immediate Westminster churn, laid out a crucial path forward. She’s advocating for a coalition of ‘middle powers’ – think fellow democracies like Japan, South Korea, Canada, and Oceania – to co-develop a resilient digital ecosystem. This isn’t about reinventing the wheel; it’s about building a sustainable, shared infrastructure that isn’t solely dependent on the ‘powerful, unaccountable few’ in Silicon Valley.
This echoes sentiments from Canadian Prime Minister Mark Carney at Davos, who called for a strategic alliance of law-abiding, middle-ranking powers to counter the swagger of authoritarian behemoths. It’s a pragmatic approach: build collective strength in areas where individual nations might be outmatched. It’s a recognition that while the US might be the dominant AI player, it doesn’t have a monopoly on innovation or a divine right to dictate the terms of the future.
The UK’s own tech secretary, while perhaps not a firebrand orator, has identified the core architectural challenge: the concentration of power. For decades, tech innovation has been characterized by rapid decentralization followed by inevitable consolidation. AI, with its colossal data and computational requirements, is accelerating this consolidation. The question for Britain isn’t just about regulation; it’s about fundamental infrastructure ownership and influence.
The Ghost of Geopolitics Past: A Familiar Pattern?
This isn’t entirely new territory, is it? Britain has a long history of finding itself reliant on a dominant global power – first the British Empire, then the post-war US alliance. But this time, the power being ceded isn’t territorial or military in the traditional sense; it’s cognitive. It’s the ability to process information, to innovate at the bleeding edge, to define the very capabilities of future machines.
Former ambassador Peter Mandelson, in his pre-sack musings, argued Britain was destined to align with Washington in a world carved into Chinese and US spheres of digital influence. Kendall’s vision offers a compelling alternative: a multi-polar digital world, where like-minded democracies can forge their own destinies. It’s a bolder, arguably more dignified, proposition than simply being a digital province of a future American technocracy.
AI sovereignty isn’t just a catchy phrase; it’s about ensuring Britain can shape its own future, rather than having it dictated by the algorithms and interests of a handful of US corporations or a volatile geopolitical landscape. The path ahead is complex, demanding strategic investment, international cooperation, and a healthy dose of skepticism towards the narratives spun by those who stand to benefit most from the status quo. The AI arms race is on, and Britain can’t afford to be a spectator.
