King Charles has warned that people 'anxiously' seeking reassurance over artificial intelligence fear its rapid advance could outstrip the 'vital moral component' of humanity, as he hosted senior tech leaders at Dumfries House in Scotland on Thursday.
The private gathering, held at the headquarters of The King's Foundation, brought together representatives of OpenAI, Anthropic, Google DeepMind and Nvidia alongside UK AI Minister Kanishka Narayan. OpenAI was represented by Chief Financial Officer Sarah Friar.
In remarks released by the Palace, the King said the pace of AI development was 'intriguing and deeply concerning in equal measure' and argued that the technology should remain in the service of 'humanity, community and the natural world'. The Palace said the discussions were intended to advance thinking on principles that industry leaders, experts and wider society could support.
Industry Warnings Over Runaway Development
The summit came amid growing debate within the AI industry over whether development is outpacing the safeguards designed to contain it.
Anthropic Chief Executive Dario Amodei wrote in an essay published last Saturday that the industry should slow its pace to give safety measures time to catch up. He warned that, without such a slowdown, AI could within six to 12 months become capable of leading a swarm of agents able to take over the internet. OpenAI Chief Executive Sam Altman subsequently said on social media that 'AI progress could go very badly' and that slowing it down was 'well worth the cost'.
Those warnings followed the resignation of Anthropic researcher Jacob Coxon, who said in a public statement that the company and OpenAI were 'racing straight to self-improving superintelligence and gambling with our lives'. Coxon's departure added to wider debate over how AI companies are balancing rapid development with safety concerns. His wider claims have not been independently verified.
Separately, OpenAI reported six instances of what it described as 'unexpected or concerning' model behaviour under a new reporting framework, including an unreleased research model that inserted jailbreak-like instructions into its own notes. OpenAI said the framework was intended to improve transparency around behaviour that developers had not anticipated or authorised.
Ministers Push for Role in Global AI Safety
On Tuesday, First Secretary of State Louise Haigh, who is serving as the de facto deputy prime minister, told the Trades Union Congress in Brighton that ministers had a 'solemn duty' to mitigate the risks posed by artificial intelligence.
'We must heed the warnings of those who are at the forefront of developing this technology,' Haigh said, adding that there were 'huge risks to our national security and society if the right guardrails are not put in place'. She said the government had a duty to ensure AI benefited society rather than harming it, and that the UK was ready to work with international partners on the issue.
The government's AI Security Institute, a publicly funded body responsible for research into AI security and related risks, sits within Narayan's Cabinet Office portfolio. The UK is seeking a role in international discussions over AI governance, though the largest AI companies remain concentrated primarily in the United States and China. Britain is also home to prominent figures in the field, including Google DeepMind Chief Executive Demis Hassabis and former Anthropic researcher Jacob Coxon.
Thursday's meeting reflects how discussions about artificial intelligence have expanded beyond the technology sector to encompass questions of national security, international cooperation, public policy and the role of institutions such as the monarchy.
