King Charles Pressed Nvidia, OpenAI and Anthropic Leaders on AI Safety at a Private Scotland Summit
The British monarch hosted a private gathering on September 18 to push some of the most powerful figures in AI toward shared safety principles. Regulators are divided, the technology is moving fast, and a royal summit cannot close that gap on its own.
Key points
- King Charles III hosted senior leaders from Nvidia, OpenAI, Anthropic and Google DeepMind at a summit in Scotland on Thursday, September 18, 2026.
- The king warned attendees that people "are anxiously seeking your reassurance that we will not lose control of our destiny," according to speech excerpts shared with CNBC.
- The gathering was convened by the King's Trust, the King's Foundation alongside the Sustainable Markets Initiative, and the U.K.'s AI minister, Kanishka Narayan.
- The summit follows rare public agreement among AI chiefs, including Sam Altman and Dario Amodei, on the need to slow development, while figures like Mark Zuckerberg and Jensen Huang have resisted.
- No binding rules were announced.
King Charles III gathered some of the biggest names in artificial intelligence, the technology that powers tools like ChatGPT and Claude, at a private summit in Scotland on Thursday. Senior representatives from Nvidia, OpenAI, Anthropic and Google DeepMind, the research lab owned by Alphabet, were in the room.
Charles did not soften his message. Buckingham Palace shared speech excerpts with CNBC, in which the king told the room: "Those in our world who value our humanity and its vital moral component are anxiously seeking your reassurance that we will not lose control of our destiny." He added that decisions taken now "will shape the world inherited by future generations."
That is a blunt thing to say to the people who control the technology in question.
What was actually on the table?
The summit focused on two practical questions: whether AI can benefit society and the natural world, and whether companies can agree on shared principles to develop it safely. U.K. AI minister Kanishka Narayan attended, lending the event some government weight. Britain has no comprehensive AI legislation in place, so actual regulatory power sits elsewhere.
Why does this moment feel different?
Because the people calling for caution are now the people building the technology. Sam Altman of OpenAI, Dario Amodei of Anthropic and Google DeepMind founder Demis Hassabis have all publicly supported slowing development and adding oversight. That position would have been nearly unthinkable two years ago. We reported on 17 September that the three firms had been coordinating quietly on safety standards for weeks, which makes this summit feel less like a surprise and more like the next step in a deliberate sequence. Our coverage of Amodei's public disagreement with Zuckerberg and how every major AI leader responded to Amodei's slowdown essay shows how sharply that debate has moved in the past week alone.
On the other side, U.S. President Donald Trump, along with Zuckerberg and Huang, has pushed back against more regulation, arguing it would damage American competitiveness.
| Attendee organisation | Position on AI regulation |
|---|---|
| OpenAI (Sam Altman) | Supports slowdown and oversight |
| Anthropic (Dario Amodei) | Supports slowdown and oversight |
| Google DeepMind (Demis Hassabis) | Supports oversight |
| Nvidia (Jensen Huang) | Opposes heavy regulation |
| Meta (Mark Zuckerberg) | Opposes heavy regulation |
A monarch hosting a safety summit changes none of that arithmetic. It does signal that the conversation has moved well outside Silicon Valley.
Charles previously delivered virtual remarks at the U.K.'s AI Safety Summit in 2023, calling AI "one of the greatest technological leaps in history of human endeavour." His tone now is noticeably more urgent.
Honest read: a summit without enforcement power is a photo opportunity with better catering. What matters is whether Altman, Amodei and the ministers present translate the language of "shared principles" into something a regulator can use. The principles keep arriving; the binding rules keep not arriving.
The one practical thing to watch: if you work in an industry AI is touching, a sitting monarch entering this conversation tells you the policy window is genuinely open. Track what the U.K. AI ministry does in the weeks after this meeting, not what was said inside it.
