Microsoft AI chief Mustafa Suleyman urges US to prioritize AI safety over China rivalry

Mustafa Suleyman has a message for Washington: stop hiding behind China. The Microsoft AI CEO appeared on CNBC’s Squawk Box on September 18 to make a case that would have sounded radical from a Big Tech executive even two years ago. The US needs AI safety regulation, he argued, and framing the conversation around competition with Beijing is just a convenient way to avoid doing the hard work. The case for treating AI like what it is Suleyman’s argument boils down to something deceptively simple. AI is a high-impact technology, and high-impact technologies get regulated. He drew parallels to other domains where powerful capabilities demanded governance frameworks, echoing historical precedents like nuclear safety. His comments carried extra weight because they arrived alongside a new corporate manifesto. Between September 14 and 16, Microsoft unveiled what it calls the “Humanist AI Code of Conduct,” a document that essentially stakes out the position that AI should remain subordinate to human interests and agency. OpenAI’s alarming disclosures and the industry divide Suleyman didn’t make his case in a vacuum. His remarks landed in the middle of an industry reckoning triggered by Open...