Microsoft president says milk faces more regulation than AI

Microsoft's President Brad Smith has argued that the technology of artificial intelligence requires more stringent global regulation compared to everyday products like milk. Speaking to the BBC, he insisted that AI currently operates under fewer safety requirements. Smith proposed a comprehensive set of safety rules, which would include mandatory disclosure to victims in the event of a system breach by a rogue AI agent, as well as the availability of an emergency brake.

He also suggested that companies should be legally obligated to submit incident reports. Smith noted that these procedures are already in place for common products such as washing machines and milk, which are subject to safety rules and inspections. The tech giants Anthropic and OpenAI have started disclosing any undesirable behavior from AI agents, contributing to a global conversation about this rapidly evolving technology.

Some companies have called for a slower pace of AI development, but this has sparked debate over potential motives, such as preserving the lead of their firms or preempting a wider government backlash. Smith suggested that the UK, with its AI Security Institute established under former Prime Minister Rishi Sunak, is well-positioned to lead in regulating AI.

The institute has pre-deployment access to major AI models, allowing it to assess their capabilities and risks. The UK's AI Minister Kanishka Narayan indicated that Britain would consider regulating advanced models if the current system proves inadequate.