Skip to main content
Microsoft President Brad Smith speaks at a podium, UK flag and AI regulation keywords in background.

Editorial illustration for Microsoft President Says UK Leads on AI Regulation

Microsoft President Says UK Leads on AI Regulation

• 3 min read

Brad Smith thinks your carton of milk gets more regulatory scrutiny than the AI models reshaping global security. Microsoft's president made that comparison on a recent BBC podcast, arguing that consumer goods like washing machines face more oversight than the companies building artificial intelligence. He called AI a "weapon and a tool" simultaneously, one that's straining the world's cybersecurity, and said the time for global safety rules is now.

Smith's pitch for who should lead that effort is specific: the UK, not Washington or Brussels. His reasoning centers on Britain's AI Security Institute, launched in 2023, which gets early access to frontier models from US labs before they go public and has built a reputation for safety testing. That access has faced pressure lately.

President Trump has criticized what he calls "globalist" organizations meddling with American firms, and reportedly pushed Anthropic and OpenAI to hold back new models from British testers. AISI said Monday it still has access to the latest systems.

Smith's comments land alongside warnings from Anthropic's Dario Amodei and OpenAI's Sam Altman, both urging a slower pace of development following a string of incidents.

Microsoft President Brad Smith is calling for urgent regulation of AI, and his argument is that consumer products like milk and washing machines have more oversight than the companies building artificial intelligence.

Why this matters

Smith's pitch is convenient. Microsoft has spent years lobbying regulators while shipping products faster than any oversight body could track, and now its president wants the UK, a market where Microsoft already has deep government ties, to write the rulebook for everyone else. That's not neutral advice, it's positioning.

The milk-and-washing-machines comparison is a good soundbite, but it also flattens a real problem: AI systems don't fail the way appliances do, and a recall notice doesn't help once a model has been deployed across millions of workflows. For builders, the signal worth watching isn't the UK's regulatory ambition, it's which jurisdiction ends up setting the compliance bar everyone has to clear. If London moves first and Washington or Brussels follow its framework, that reshapes where startups incorporate, where they test agents, and how much legal cover "going fast" still gives them.

Amodei and Altman calling for brakes while their companies keep shipping is its own tell. Watch who actually writes binding rules, not who talks about wanting them.

LIVE20:35Sakana AI Model Improves Problem-Solving but Retains Blind Spots