Brad Smith, President of Microsoft, has emphasized the urgent need for regulatory measures concerning artificial intelligence (AI), arguing that consumer goods such as milk and washing machines are subject to more stringent oversight than the organizations developing AI technologies. In a recent podcast with the BBC, Smith characterized AI as both a “weapon and a tool” that poses significant risks to global cybersecurity, asserting that the “moment has arrived” for the establishment of international safety regulations.
Smith further stated that the United Kingdom is uniquely positioned to spearhead the development of these global safety standards, rather than the United States or the European Union. The AI Security Institute in Britain, which was established in 2023, has gained access to major AI models from leading U.S. laboratories prior to their deployment, quickly establishing itself as a frontrunner in AI safety assessments.
However, this position has faced challenges in recent weeks, particularly following remarks from U.S. President Donald Trump, who has cautioned against “globalist” entities that he claims are undermining American businesses. Reports indicate that last month, Trump urged companies such as Anthropic and OpenAI to refrain from sharing new models with British evaluators. Nevertheless, the AISI confirmed on Monday that it retains the ability to access and evaluate the latest AI models.
Despite the president’s objections, Smith contended that the U.K. possesses the capability to regulate AI in ways that the U.S. cannot. “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,” he remarked during the BBC interview. “That’s a strength.” His statements coincide with growing calls from technology leaders, including Dario Amodei of Anthropic and Sam Altman of OpenAI, to decelerate AI development following several incidents involving errant AI behavior.
Last month, Trump convened tech executives from companies such as OpenAI, Anthropic, Nvidia, AMD, SpaceX, Meta, and Google at the White House to discuss AI safety. They collectively endorsed a “morally binding” agreement to collaborate with “independent auditors” in evaluating the safety and functionality of their AI models.
Smith described this agreement as a “good step,” but he believes it falls short of the comprehensive regulation required. He expressed concern that voluntary self-regulation would only be adhered to by “good companies.” “It’s the companies that are not as responsible that you need to worry about,” he stated.
Smith proposed several AI safety regulations he would like to see implemented, including a legal requirement for organizations to be notified if their systems are compromised by rogue AI agents. He also expressed support for AI laboratories obtaining approval from external evaluators prior to releasing AI products and advocated for the inclusion of kill-switch mechanisms to halt or slow AI operations if human oversight is lost.
“Let’s speak calmly, maybe even softly, but firmly, and start to bring people together,” Smith urged. “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 concluded.
