Brad Smith, President of Microsoft, has called for urgent regulation of artificial intelligence that is grounded in law. Speaking on a recent BBC podcast, he argued that consumer goods such as milk or washing machines today face more oversight than the companies building AI, and described AI as both a “weapon and a tool” that is putting global cybersecurity at risk.
Why action is needed now
Smith said "the moment has arrived" to put global safety rules in place. He warned that voluntary self-regulation is insufficient: while responsible companies will comply with voluntary measures, the real danger comes from less responsible actors who would not.
Why the UK could lead
Smith argued that the United Kingdom, rather than the United States or the European Union, is well positioned to lead on global AI safety standards. He pointed to Britain’s AI Security Institute (AISI), launched in 2023, which gains access to major models from U.S. frontier labs before public deployment and has quickly become a leader in AI safety testing.
Political friction and access concerns
The AISI’s role has been threatened recently after U.S. President Donald Trump criticized so-called “globalist” organisations and, according to reports, asked Anthropic and OpenAI last month to withhold new models from British testers. AISI confirmed on Monday that it is able to access and test the latest models despite that pushback.
Smith said part of the UK’s advantage is perception in Washington: "In Washington, D.C., people don’t look at the U.K. and they don’t look at London the way they look at Brussels. That’s a strength," he told the BBC.
Specific safety measures Smith proposes
Smith outlined several concrete rules he would like to see enacted into law, including a legal obligation to notify an organisation if its system has been breached by rogue AI agents. He also favoured requiring AI labs to obtain approval from external evaluators before releasing products, and spoke positively about the use of kill-switches to pause or slow AI when humans lose control.
Industry steps and limitations
Smith referenced recent gatherings of tech leaders to discuss AI safety, including a White House meeting last month attended by executives from OpenAI, Anthropic, Nvidia, AMD, SpaceX, Meta and Google. Those companies signed a "morally binding" document agreeing to work with independent auditors to assess their models’ safety and capabilities. Smith described the document as "a good step," but said it falls short of the mandatory regulation that he believes is necessary.
Call for calm, firm international cooperation
Smith urged stakeholders to begin calm, firm efforts to bring people together around enforceable rules. "I hope that’s the role we can play as a company. I hope that’s the role the U.K. can play as a country," he said to the BBC.
The remarks underline growing calls for internationally coordinated, legally enforceable AI governance — particularly around external evaluation, breach notification, and fail-safe mechanisms — as AI systems become more powerful.
