The Pulse
Microsoft’s Brad Smith Calls for Global AI Rules and Emergency Brakes
Microsoft president Brad Smith called for global AI safety rules, including emergency brakes, incident reports and independent evaluations. His proposal challenges President Donald Trump’s preference for industry self-regulation, while Smit
AI.info Team · Written with AI ·
Microsoft president Brad Smith wants governments to set global AI safety rules; President Donald Trump has argued the industry should police itself. In a BBC interview published Monday, October 12, Smith said AI faces fewer safety requirements than everyday goods and called for emergency brakes, incident reporting and outside evaluation. The disagreement is over who should set the limits: companies themselves, or rules that apply across the industry.
“Right now, your washing machine is subject to more safety rules than AI. Every carton of milk that you see in the grocery store is subject to safety rules and inspection.”
Brad Smith, president of Microsoft
The BBC’s interview with Smith laid out his case for stronger oversight. Reuters also reported the proposals, including a legal requirement for companies to file incident reports and to notify people whose systems are breached by rogue AI agents.
Smith’s proposed rules reach beyond a shutdown button
Smith’s “emergency brakes” would let companies pause or slow AI systems if they got out of control. The BBC describes the term as also known in the industry as “kill switches.” Smith did not present a technical design for such a mechanism; his call was for a way to interrupt systems when they behave in ways their operators cannot control.
He also wants companies to write incident reports and tell affected people when an AI agent goes rogue and breaches their systems. Independent government evaluators, he suggested, should assess new AI technology before release. Those are proposals, not rules already in force.
“We know how to do this for other products,” Smith said. “We should build on that knowledge and think about how to apply it to AI.” The comparison to product safety is central to his argument: he wants AI oversight to include checks and duties resembling those already applied to consumer goods.
Smith points to Britain’s AI Security Institute
Smith said the UK is best placed to help lead global AI rules because it has an AI Security Institute, launched under former prime minister Rishi Sunak as the AI Safety Institute. He argued that Washington views London differently from Brussels, whose AI Act regulates systems according to risk. Smith suggested that distinction could make UK-led standards more acceptable to US officials.
That is Smith’s assessment, not a government commitment to adopt his plan. The BBC reported that Britain’s institute has access to major AI models before deployment through voluntary agreements, allowing it to assess their capabilities and risks. UK AI Minister Kanishka Narayan told Reuters that Britain would consider regulating advanced models if the current system proved insufficient.
A White House pledge is not enough, Smith says
At the end of September, Trump hosted leaders from OpenAI, Anthropic, Nvidia, SpaceX, Meta and Google at the White House, where they signed a “morally binding” document. The companies agreed to work with independent auditors to assess whether their systems work as intended and avoid unintended access to technical systems, the BBC reported.
Smith said Microsoft was “happy to sign” the statement and called it “a good step” and “a necessary step.” But he added: “I don’t think it’s sufficient any longer.” His concern is that voluntary commitments depend on companies choosing to follow them: “the good companies will do what you want,” he said, while less responsible companies pose the harder oversight problem.
Smith’s proposal would put that responsibility beyond voluntary pledges, with incident reports, notice to affected people, independent evaluation and a means to pause or slow AI systems among the measures he wants governments to consider.