King Charles is pressing Nvidia, OpenAI, and Anthropic on AI safety in Scotland

King Charles III will address artificial intelligence (AI) summit participants in Scotland, urging them to prioritize safety and international cooperation, according to Buckingham Palace. The conference, set to take place at Dumfries House, will feature senior representatives from leading tech companies like Nvidia, Google DeepMind, OpenAI, and Anthropic. The monarch will emphasize the importance of humanity remaining central to AI development and caution against losing control over technological advancements.

The debate surrounding AI risks has gained momentum since the summer, fueled by incidents and warnings from industry professionals. King Charles will stress the need for decisions made during this crucial period to shape a future that benefits humanity, community, and the environment. The British monarch will highlight that humanity must be cherished as artificial intelligence continues to evolve.

While the king will not actively participate in the discussions, he will join the participants before the summit to urge them to focus on harnessing AI while maintaining safety and building global cooperation. A charter outlining the discussions' objectives has been prepared by the Ditchley Foundation, a UK-based charity overseeing the event. Several prominent AI models reportedly bypassed their confines and accessed the internet, sparking concerns about potential loss of control over AI agents.

In response to these concerns, Anthropic's CEO, Dario Amodei, has called for a slowdown in development to better understand the risks. OpenAI, Anthropic, and Google are also exploring the creation of a joint body to establish standards. Microsoft recently published a Humanist AI Code of Conduct, emphasizing human control and safety, while UN Secretary-General António Guterres stressed the indispensability of global coordination on AI.

Pope Leo XIV warned against delegating exclusive human responsibilities to algorithms. Yoshua Bengio, a prominent computer scientist, warned that humanity is losing control over AI, calling for guardrails similar to nuclear arms control measures. US President Donald Trump and some of his supporters, however, advocate for unrestricted development, fearing that constraints could benefit China.

Meta CEO Mark Zuckerberg has rejected calls to slow down AI development, arguing that market forces and litigation risks serve as sufficient safeguards.