TL;DR
The King used Thursday’s summit at Dumfries House to warn bosses from OpenAI, Nvidia, Google DeepMind and Anthropic about the “existential dangers” of AI, asking whether “we need sufficient means of control before it is all too late”. The intervention lands a week before Prime Minister Andy Burnham meets Donald Trump for the first time, and one MP thinks the timing may help him.
What the King said
Speaking at the King’s Foundation headquarters in East Ayrshire, Charles called the speed of AI progress “deeply concerning” and said development must be in the “service of humanity, community and the natural world”. He pointed out that the people building these systems are themselves now warning of “darker capacities”, and said there was “urgency in adequately considering the existential dangers of such technologies falling into the wrong hands”.
According to the Royal Household, the Ditchley Foundation ran the discussions. Delegates considered whether a common set of principles could guide how AI is applied, framed around human dignity as well as efficiency. AI minister Kanishka Narayan attended, alongside representatives of the King’s Trust and the Sustainable Markets Initiative, the climate network the King founded. Pope Leo XIV’s adviser on technology ethics, Father Paolo Benanti, attended for the Vatican.
Tino Cuéllar, who runs global affairs at Anthropic, said the firm was “honoured” to take part. In the run-up, Microsoft AI’s Mustafa Suleyman went on the BBC to criticise Anthropic’s approach.
Why the timing matters
A monarch has no power to regulate, and yesterday’s preview made that point. What changes the reading is the diary. Burnham is in New York next week for the UN General Assembly, where he will meet Trump face to face. Trump has so far pushed back hard against tighter AI rules, and has a well-known fondness for the monarchy.
Labour’s Dame Chi Onwurah, chair of the Commons committee on science and technology, said the King’s “usual quiet diplomacy on this critical subject should be helpful to the prime minister in his conversations with Trump”.
The summit also follows a busy fortnight. Anthropic’s Dario Amodei called for a slowdown that Elon Musk and OpenAI’s Sam Altman then endorsed, and more than 70 parliamentarians asked Burnham to back a ban on superintelligence. The King’s words give the Prime Minister cover to raise safety with Trump directly.
Looking forward
The test is what Burnham says in New York. If he brings AI control into his first Trump meeting, the Dumfries House speech will look like the opening move in a UK position. If safety stays off the agenda, the summit will be remembered as a well-staged conversation with no follow-through. Watch too for any published version of the principles delegates discussed, which would show whether the four labs signed up to anything specific.