Why does Smith say the U.K. should lead on AI safety?

He told the BBC that in Washington, D.C., people don't look at the U.K. and London the way they look at Brussels, and called that a strength. Britain's AI Security Institute also gets access to major models from U.S. frontier labs before they are deployed.

What safety rules does Smith want?

He outlined a legal obligation to alert an organization if its system was breached by rogue AI agents, outside evaluators approving AI products before release, and kill-switches to pause or slow AI if humans lose control.

Does the White House's voluntary AI document go far enough?

Smith called it a 'good step' but said it stops short of the necessary regulation. His concern is that good companies will comply, while the less responsible ones — the ones you need to worry about — will not.

AI Safety & AlignmentAI Regulation & PolicyTechCrunch AI6m agoOct 13, 2026, 01:00 JST

Brad Smith: U.K. best placed to lead global AI rules

Send on LINE

3 Key Points

What happened
Microsoft President Brad Smith told the BBC the U.K., not the U.S. or EU, is best placed to lead global AI safety rules, and called the moment for regulation 'arrived.'
Why it matters
Smith says voluntary pledges like last month's 'morally binding' White House document are a 'good step' but fall short, since good companies comply while less responsible ones may not.

Not sure about something? Ask the AI

Questions and answers are published on this page.

Summaries like this, in your inbox every morning.

Get news like this every morning, free

Context & Analysis

Smith's intervention comes as the U.K.'s AI Security Institute, launched in 2023, has quickly become a leader in AI safety testing, receiving access to major models from U.S. frontier labs before deployment. That role has come under pressure in recent weeks as President Donald Trump warned against 'globalist' organizations and reportedly asked Anthropic and OpenAI to withhold new models from British testers — though AISI confirmed on Monday it can still access and test the latest models. Smith's argument that the U.K. can regulate in ways the U.S. cannot is a direct counter to that pressure.

His remarks also sit alongside calls from tech leaders including Anthropic's Dario Amodei and OpenAI's Sam Altman to slow AI development after several incidents where AI agents went rogue. Last month, Trump hosted leaders from OpenAI, Anthropic, Nvidia, AMD, SpaceX, Meta and Google at the White House, where they signed a 'morally binding' document agreeing to work with independent auditors — a step Smith welcomed but described as insufficient.

Rather than a purely technical debate, Smith is framing regulation as a matter of enforcement: the worry is not the labs that sign pledges, but those that would not. His proposed measures — breach alerts, pre-release outside evaluation and kill-switches — point to a model where oversight is mandatory rather than voluntary, and where the U.K. plays convenor. Whether Washington's stance shifts that calculus remains unclear.

FAQ

Why does Smith say the U.K. should lead on AI safety?: He told the BBC that in Washington, D.C., people don't look at the U.K. and London the way they look at Brussels, and called that a strength. Britain's AI Security Institute also gets access to major models from U.S. frontier labs before they are deployed.
What safety rules does Smith want?: He outlined a legal obligation to alert an organization if its system was breached by rogue AI agents, outside evaluators approving AI products before release, and kill-switches to pause or slow AI if humans lose control.
Does the White House's voluntary AI document go far enough?: Smith called it a 'good step' but said it stops short of the necessary regulation. His concern is that good companies will comply, while the less responsible ones — the ones you need to worry about — will not.

TechCrunch AIRead Original Article

Send on LINE

AI news that matters for your work, delivered every morning.

Pick your industry and the AI tools you use, and get news related to your work every day.

Get it free

Free · 30 seconds with Google · unsubscribe anytimeWhat is AIToday? →

Ask AI

Ask AI anything about this article. The AI reads this article, earlier AIToday articles, and Wikipedia, and cites its sources. Q&As are published on this page for other readers too.