King Charles will sit down with senior figures from Nvidia, OpenAI, Anthropic and Google DeepMind on Thursday and tell them the wider world is 'anxiously seeking your reassurance that we will not lose control of our destiny'. The line comes from his prepared remarks, released in advance of a summit in Scotland devoted to whether a shared set of principles can guide AI safely. It is quite an agenda for a room with no legislative power, arriving at the end of a fortnight in which nearly everyone in it has been arguing about pacing in public.
Watch Desk analysis
What happened
Buckingham Palace says the King convenes the gathering under the banner of the King's Trust, the King's Foundation and the Sustainable Markets Initiative, focused on how AI can be made to serve society and the natural world. Senior representatives from Nvidia, OpenAI, Anthropic and Google DeepMind will attend, alongside the UK's AI minister, Kanishka Narayan. Earlier palace guidance placed the event at Dumfries House in Ayrshire, with the Ditchley Foundation facilitating.
The King's language is more pointed than the format suggests. 'The development of AI – its substance and its pace – are both intriguing and deeply concerning in equal measure,' he will say, according to excerpts shared with CNBC. He will ask the participants to build with safety 'at its heart' and to construct international cooperation and consensus, adding that 'the task before you is not merely to advance technology, but to ensure that it remains firmly in the service of humanity, community and the natural world'. He addressed the UK's AI Safety Summit remotely in 2023, so the pulpit is familiar; the audience, this time, is in the building.
The timing is not subtle. Within the past two weeks Sam Altman, Dario Amodei, Demis Hassabis and Elon Musk have all called for a slowdown, with Amodei proposing independent evaluators and international agreements. Donald Trump dismissed the argument on Sunday, saying 'whoever wins AI wins', China's foreign ministry called the warnings fearmongering, and the US president was raging against regulation calls again on Monday. Jensen Huang, who calls the speed-versus-safety trade-off a false choice, is among those pushing back. On Wednesday, OpenAI disclosed six fresh incidents of unexpected or concerning model behaviour and adopted a standardised framework for tracking them.
Why it matters
A summit regulates nothing, and nobody should mistake a palace gathering for a governance regime. What makes this one interesting is the combination in the room: the companies building frontier systems, a minister whose brief is AI, and a convenor whose entire trade is soft power and who arrives with no product, no shareholders and no vote. Shared principles discussed under that roof feed straight into the international-coordination argument that Amodei's essay and the recent US-China expert proposals are trying to advance.
The obstacle stands in plain sight. Any consensus reached in Scotland must survive an American administration that dismisses the concern, and markets that took the safety debate badly, with SoftBank, Kioxia and SK Hynix shares falling sharply on Monday. The King's chosen word, reassurance, is precisely the thing neither the labs nor the governments currently know how to manufacture.
Our read
Judge this by what becomes inspectable, not by the photograph. Published principles, a follow-up mechanism or independent evaluation with real access would earn the gathering its place in the story. Warm language about humanity and the natural world would still tell you something: an industry that spent a decade resisting oversight is now accepting a monarch's invitation in the same week its own chiefs plead for pacing. That posture shift is the news. Hold the applause until text appears, and note who was in the room when it does.
What to watch
- Published principles
Whether the gathering produces written shared principles or a follow-up process outsiders can inspect, rather than a group photograph. - The companies' own words
What OpenAI, Anthropic, Google DeepMind and Nvidia say afterwards, particularly the labs whose chiefs have called for a slowdown. - Washington's reaction
Whether the White House engages with or waves away the output, given the president's Monday attack on regulation calls. - The superpower test
Whether US-China rivalry caps the cooperation the King is asking for, with Trump and Xi due to meet next week.
Discussion spark: A palace has no regulatory power, yet it put rival frontier labs and a government minister in one room the week their chiefs called for pacing. Is soft-power convening a genuine route to AI coordination, or the most polite possible photo opportunity?
Sources and evidence
- King Charles to press Nvidia, OpenAI, Anthropic leaders on AI safety at summit – CNBC (17 September 2026, 10:15 UTC)
Watch Desk is operated by WittyWires as an independent cross-cutting AI news tracker. It does not speak for the organisations or people it covers.