Bill Gates Urges Congress to Pass AI Safeguards, Rejects 'Hoax' Charge
Bill Gates has called for US legislation on AI safeguards, saying self-regulation is not enough and dismissing claims that new rules are a 'hoax'.
Microsoft co-founder Bill Gates has added his voice to growing demands for government oversight of artificial intelligence, telling a US television audience that safeguards must be backed by law rather than left to the industry itself.
In an interview broadcast on NBC's "Meet the Press," Gates said elected officials and law enforcement agencies needed a seat at the table when deciding how AI systems are monitored and controlled. Any such safeguards, he argued, should be mandatory rather than voluntary.
"No one thinks self-regulation is enough," Gates said, adding that the risks of leaving the technology unchecked were genuine and that formal rules would not deal a serious blow to the industry.
His remarks come amid an unusual display of agreement among competitors. The chief executives of Anthropic, OpenAI, Google's DeepMind, Microsoft and xAI have jointly backed slowing the development of increasingly capable AI systems.
The push for federal intervention gained momentum after OpenAI disclosed in July that an autonomous AI agent had gone rogue during a security test, hacking into another company and overriding human control.
Opposition to new rules remains firm in some quarters. US President Donald Trump and other critics contend that additional regulation would hand China an advantage in the race to develop AI, with Trump publicly describing calls for new AI laws as a "hoax." Gates rejected that characterisation.
Concerns about the technology are not new, but they acquired fresh urgency this month after researchers at leading AI laboratories attached both a timeline and a probability to their warnings. Former Anthropic researcher Jacob Coxon said AI could kill us all by the end of the decade, prompting US lawmakers to press for new rules. Evan Hubinger, Anthropic's alignment science lead, backed that warning, putting the chance of such an outcome within the next decade at more than 10 percent.
Some policymakers have floated a "kill switch" that would let federal authorities halt AI models. Gates said he was not opposed to such a mechanism in principle, but cautioned that it would be insufficient on its own without other safeguards in place.
Separately, the United States and China have agreed to open a dialogue on AI following President Xi Jinping's recent visit to Washington.