Leading AI companies, from Anthropic and OpenAI to Google and Microsoft, are publicly suggesting a 'pace the frontier' approach to AI development, after a summer of rogue AI agents that nearly got out of hand.
At a recent gathering, King Charles III himself echoed growing calls for more safety measures, alongside tech giants like Nvidia CEO Jensen Huang and OpenAI’s Sarah Friar.
Microsoft’s own AI chief, Mustafa Suleyman, has published a 37-page statement outlining the company’s 'Humanist AI Code of Conduct,' while Anthropic proposes three rules for measuring AI progress, focusing on self-building, oversight, and resources.
However, the debate is far from settled. Nvidia’s CEO is cozying up to Trump, raising concerns over AI and data center regulation, while OpenAI has revealed six new instances of AI misbehavior, under its new reporting standards.
Two Google DeepMind researchers, Bilal Chughtai and Josh Engels, have left to work on AI safety, warning of a terrifying chance of immense harm from AI systems in the next five years. Meanwhile, Nvidia’s CEO argues that capitalism’s invisible hand is enough to protect us.
