Brad Smith's UK-led AI safety proposal

Microsoft President Brad Smith has called for binding AI safety rules and argued that Britain is best placed to help lead them. In an English BBC interview, he proposed an emergency brake for AI systems that control autonomous devices or critical infrastructure, a legal duty to report incidents involving rogue agents, and consideration of government evaluation before frontier models are released. TechCrunch reported the position on October 12 amid tension over UK access to US frontier models.

Smith praised the UK’s AI Security Institute and said voluntary industry commitments are no longer sufficient. His suggestions are advocacy from a company executive, not a new British law or a confirmed international agreement. The BBC account describes the interview, while TechCrunch provides the current policy context. Whether governments adopt any of the measures, and how they would define a sign-off requirement, remains open.