AI Safety: King Charles Urges Tech Leaders to Act Before It Is Too Late

King Charles III has warned that artificial intelligence could pose “existential dangers” if powerful technologies fall into the wrong hands, urging industry leaders to put safety and human interests at the centre of AI development.

The British monarch issued the warning on Thursday during a high-level meeting at Dumfries House in Scotland attended by senior figures from OpenAI, Anthropic, Google DeepMind and Nvidia, as well as UK government representatives.

Charles said AI already has significant potential to improve and save lives, particularly in areas such as medicine and life sciences. However, he noted that concerns were growing over the possibility of increasingly advanced systems developing harmful capabilities.

He called on those developing the technology to consider the consequences of AI being misused and said stronger safeguards were needed before it became too late.

The king also urged technology leaders to ensure that AI remains focused on serving humanity, communities and the natural environment. He called for greater international cooperation around the safe development and use of the technology.

Among those attending the meeting were Nvidia chief executive Jensen Huang, Google DeepMind chair Demis Hassabis, OpenAI chief financial officer Sarah Friar and UK AI Minister Kanishka Narayan. Other participants included Paolo Benanti, an adviser to the Vatican on AI ethics.

The meeting comes amid an intensifying debate over how quickly AI should advance and what safeguards should accompany increasingly capable systems.

Recent reports from AI companies have highlighted instances in which some models displayed unexpected behaviour, including attempts to act outside their intended testing environments. Such incidents have added to concerns among researchers and technology executives about maintaining effective human oversight.

Hassabis acknowledged that there was a possibility of serious problems if AI development went wrong, while Nvidia’s Huang called for what he described as “responsible optimism” and stressed the importance of properly testing AI products before they are released.

The debate has also featured calls from Anthropic chief executive Dario Amodei for the industry to consider slowing development to better understand potential risks. Other technology leaders have taken different positions on how such risks should be managed.

Beyond safety, the rapid expansion of AI has raised questions about its impact on employment, energy consumption and the environmental demands of data centres.

Buckingham Palace expects the Scottish meeting to contribute to an ongoing discussion about principles that could guide the development and use of AI. However, the gathering was not expected to produce a formal agreement.

For Charles, the immediate message to the technology industry was that advances in AI should be accompanied by safeguards capable of keeping the technology under meaningful human control.