TL;DR

Microsoft president Brad Smith has told the BBC that a carton of milk or a washing machine faces more safety rules than AI, and called for global regulation to start now. In a Big Boss Interview podcast published on Monday 12 October, he named the UK as the country best placed to lead that effort, putting him at odds with President Trump’s preference for industry self-regulation.

What Smith wants

His proposals borrow from product safety. Companies should have “emergency brakes”, the industry’s kill switches, to pause or slow AI that gets out of control. They should be legally required to file incident reports and to inform a “victim” when a rogue AI agent breaches their systems. And new AI should be vetted by independent evaluators working for governments, with agreed rules on when outside sign-off is required before a product is released.

“We know how to do this for other products,” Smith said, describing AI as both “a tool and a weapon”.

He welcomed the “morally binding” statement that US tech leaders, including OpenAI, Anthropic and Google, signed at the White House at the end of September, committing to independent auditors. Microsoft would happily sign it too, he said, but self-regulation is no longer enough because “the good companies will do what you want”; it is the less responsible ones that need watching.

Why the UK

Smith credits the AI Security Institute, created under Rishi Sunak as the AI Safety Institute, with giving Britain a technical lead that, he said, most people he knows in tech would readily name. He also argued that Washington views London very differently from Brussels, suggesting the US would accept UK-made standards more readily than EU ones.

Looking forward

Smith’s remarks land alongside a similar call from his own chief executive. Satya Nadella urged on Saturday that AI models be built with an “emergency brake”, The Guardian reports, and the paper notes that leaders of countries including France, Germany and Australia have signed a call for coordinated safety standards that the UK has not. The interview went out on the same day AI minister Kanishka Narayan was due to argue that firms cannot be relied on entirely to police themselves. In our view, a senior US executive publicly casting the UK as the natural rule-setter strengthens ministers’ hand at next year’s G20, but only if they bring concrete proposals rather than principles. Smith’s endorsement of AISI may also matter for UK firms, since it hints that evaluation standards set in London could carry weight with the largest US vendors.