Microsoft president Brad Smith wants governments to treat advanced AI more like other products whose makers already face legal safety obligations. In a BBC interview, he called AI both a “tool and a weapon” and said global safety rules are urgently needed.
“Right now, your washing machine is subject to more safety rules than AI,” Smith said. “Every carton of milk that you see in the grocery store is subject to safety rules and inspection.” The comparison was meant to underline his case for a common international framework rather than to suggest that AI has no rules at all.
The safeguards Smith wants
Smith proposed several specific requirements. Companies would have to file incident reports and notify people when a rogue AI agent breaches their systems. Advanced systems would also need an emergency brake, often called a kill switch, that could pause or slow them if they moved out of control.
He also said independent government evaluators should help decide when a new AI product needs outside approval before release. Reuters reported that Smith framed these measures as an extension of safety practices already used for other products, not a wholly new regulatory concept.
The proposal goes beyond a recent voluntary US pledge signed by leaders of major AI companies. That agreement committed participants to independent auditing and safeguards designed to prevent systems from hacking or accessing technical infrastructure in unintended ways. Smith called it a good and necessary step, but told the BBC it was no longer sufficient. He warned that any self-regulatory system depends on companies choosing to comply.
Why he thinks the UK could lead
Smith argued that Britain is well placed to shape global standards because of its AI Security Institute. The institute receives access to major AI models before release through voluntary agreements, allowing it to assess capabilities and risks. That access does not currently give it a mandatory veto over product launches.
He also suggested Washington may be more receptive to standards developed in London than to the European Union’s risk-based AI Act. “People don’t look at the UK and they don’t look at London the way they look at Brussels,” Smith said, describing that difference as a potential advantage.
The political gap remains substantial. President Donald Trump has favored industry self-regulation and argued that US companies need room to compete with China. Smith is staking out a stricter position, asking governments to turn voluntary safety practices into enforceable obligations before increasingly autonomous agents cause harm.
