Microsoft president Brad Smith: AI is less regulated than milk, wants 'kill switches'

Microsoft president Brad Smith told the BBC that AI is less regulated than milk or washing machines and needs urgent safety rules, The Guardian reported on 12 October 2026. He called for "kill switches" that let companies pause or slow an AI system that grows out of control, and for a legal obligation to compensate a "victim" whose system is breached by a rogue AI agent. His comments follow Microsoft CEO Satya Nadella's call on Saturday for an AI "emergency brake". The report notes Microsoft runs AI infrastructure and is a shareholder in OpenAI. The piece adds that the UK's AI minister, Kanishka Narayan, was due to say on Monday that firms building frontier AI should do more on safety, and that the UK government is preparing to legislate for more oversight, according to reports. On Hacker News (65 points, about 70 comments), reaction split. The largest group mocked the comparison, arguing text-and-image software should not be regulated like something you ingest, with the top comment joking that operating systems are less regulated than milk too. A sizeable camp was cynical about the motive, saying a company behind in AI wants brakes applied to rivals. A few argued the risks are serious and the casual dismissals miss the point. No consensus. Sorted take: the soundbite travels, but the substance to watch is the push for kill switches and liability, which would fall on firms that deploy AI.