Bill Gates has waded into the escalating AI regulation debate with a blunt message for Washington: leave AI safety entirely in the hands of the companies building it, and you’re taking a gamble the country can’t afford. The Microsoft co-founder told NBC News’ “Meet the Press” that lawmakers need to step in now, arguing that voluntary industry oversight simply isn’t sufficient to manage the risks posed by fast-moving artificial intelligence systems.
Key takeaways
- Bill Gates said self-regulation by AI companies “is not enough” and called for lawmakers and law enforcement to help set safeguards and monitoring rules for AI.
- Gates warned that AI is powerful enough to enable events causing mass casualties if misused by people with harmful intent.
- Anthropic’s Dario Amodei and OpenAI’s Sam Altman have both pushed for slowing AI development and building international standards, including cooperation with China, while Meta’s Mark Zuckerberg opposes industrywide coordination.
- House Speaker Mike Johnson declined to bring Congress back to legislate on AI’s pace, saying he would defer to the companies themselves.
- California, Maryland and New York have each launched state-level AI regulatory initiatives as federal action stalls.
Bill Gates Says Self-Regulation Isn’t Enough — Calls for Government Oversight of AI
Gates left little room for ambiguity when moderator Kristen Welker asked whether Washington needs to pass legislation on AI. “Absolutely,” he said. “No one thinks self-regulation is enough,” Gates told “Meet the Press” in an interview set to air in full Sunday.
His argument centers on a simple idea: the people writing the rules shouldn’t be the same people racing to build the product. “You need law enforcement and the politicians to get into the discussion about what safeguards and monitoring look like,” Gates said, adding that formal oversight “has to be a required thing.” He acknowledged it would create “a little bit of overhead for the industry, but not a dramatic slowing of what they’re doing.”
Why Gates warns AI could cause “a billion deaths”
Gates didn’t dismiss longer-term fears that AI could eventually threaten humanity, but he steered the conversation toward a more immediate danger: misuse by people who already mean harm. “AI is certainly powerful enough to drive events that, you know, cause a billion deaths,” he said. “Even though it’s pretty hard to get to 100%, there’s never been a weapon as powerful as the combination of people with ill intent using the latest AI tools.”
That framing matters for the broader policy fight. Rather than treating AI risk as a distant, hypothetical scenario, Gates is describing it as a present-tense security problem — one that US AI government policy hasn’t yet caught up with.
AI Industry Leaders Are Split on How Fast — and How Far — to Regulate
The industry itself is far from unified on what comes next, and that division is arguably the most consequential part of this story. Earlier this month, the CEOs of Anthropic and OpenAI — two of the country’s most influential AI labs — agreed that the race to build increasingly powerful systems needs to slow down. That rare alignment came just as a former Anthropic employee went public with much sharper warnings.
Anthropic and OpenAI CEOs call for a slowdown
In an essay laying out his case for pacing AI development, Anthropic CEO Dario Amodei argued that regulation needs teeth and needs to be broad. “The most effective method of pacing is via regulation that targets all U.S. frontier AI companies, as that covers even those who are unwilling to cooperate voluntarily,” he wrote. He went further, stating that “global pacing will require cooperation with China, the autocratic country with by far the most advanced AI capabilities.”
Speaking before a United Nations gathering, OpenAI CEO Sam Altman echoed that call for shared rules rather than company-by-company judgment calls. He pressed for international standards for “measuring capabilities, assessing risks, determining whether safeguards are sufficient and preserving meaningful human oversight” over AI systems.
A whistleblower’s warning shakes the debate
The urgency behind these statements traces back to Jacob Coxon, a whistleblower who recently resigned from Anthropic. Coxon publicly declared that “neither company is acting responsibly. They are racing straight to self-improving superintelligence and gambling with our lives.” In a series of social media posts and interviews, he added that “the people building AI earnestly believe that it could kill us all by the end of the decade.”
Notably, Coxon’s own prescription for fixing the problem doesn’t fully align with Gates’ call for government intervention. In an earlier “Meet the Press” appearance, Coxon said he believes AI companies should regulate themselves, arguing that “the people running the labs, especially their recent communications, are completely genuine. They would like to slow themselves down.”
Zuckerberg pushes back on industrywide coordination
Not every major tech leader agrees that cross-company rules are the answer. Meta founder and CEO Mark Zuckerberg told NBC News in an interview published Thursday that he sees things differently. “I don’t think that we need some kind of industrywide coordination,” Zuckerberg said. “I think that each lab needs to take the time, and when it sees that there are issues, you just take the time that you need internally to basically make sure that you’re proceeding safely.”
That stance puts Zuckerberg at odds with Amodei and Altman, and it highlights a genuine fault line running through Silicon Valley — one where the biggest disagreement isn’t whether AI carries risk, but who should be trusted to manage it.
Microsoft President Brad Smith, speaking at a UN session on AI and human development, was far more pointed in his criticism of companies that keep pushing products out the door. He compared the situation to the commercial airline industry. “Imagine for a moment that there’s only two big companies in the world that produce commercial airplanes,” Smith said. “Imagine that somebody comes into your office and says, ‘I know we’re in a race to get out the next airplane first, but we’re worried that the airplane will crash and people will die.’ Would you say, ‘I need to win this race’? Or, would you say, ‘I’m going to put the safety of my passengers and the safety of the public ahead of my winning that race’? That is the fundamental question.”
Washington Defers, States Step In
While AI executives argue among themselves about the right framework, the federal government has largely stayed on the sidelines — leaving individual states to fill the gap. That split between inaction in Washington and activity in state capitals is quickly becoming its own storyline within the wider AI regulation debate.
Speaker Johnson leaves AI regulation to the industry
Appearing on “Meet the Press” several weeks ago, House Speaker Mike Johnson, R-La., declined to bring lawmakers back to Washington to pass legislation regulating the pace of AI growth, saying he’d rather let the companies lead. “We have to put some guardrails, some safety measures in place to ensure that AI doesn’t run away,” Johnson said, but he noted that even the CEOs calling for a slowdown “don’t themselves even have the answer as of today. So we can’t rush in and have Congress do something that would be detrimental to national security.”
California, Maryland and New York move on their own
Rather than wait for stalled federal legislation, several governors have taken matters into their own hands. Last week in California, Gov. Gavin Newsom signed an executive order establishing a board of experts charged with drafting a plan to bolster the state’s AI safety laws.
This week, Maryland Gov. Wes Moore and New York Gov. Kathy Hochul each rolled out their own plans aimed at tightening AI oversight. Moore’s order sets up what’s described as a “three-pronged approach” to regulation, while Hochul’s creates a new Office of Digital Innovation, Governance, Integrity and Trust to monitor legal compliance among AI companies operating in the state.
Taken together, these moves suggest that even if Congress keeps deferring to industry, Bill Gates AI oversight concerns are already being addressed piecemeal at the state level — raising the possibility of a patchwork of rules that vary sharply depending on where an AI company operates. Whether that fragmented approach can substitute for the federal or global framework that Amodei and Altman are calling for remains an open question, and one that’s likely to shape how the broader AI development risks debate plays out heading into next year.
FAQ
Why does Bill Gates believe government regulation of AI is necessary?
Bill Gates stated that self-regulation by AI companies isn’t enough and that law enforcement and politicians need to be involved in safeguarding and monitoring AI development.
What risks does Bill Gates highlight regarding AI misuse?
Gates warned that AI is powerful enough to cause large-scale harm, including potentially causing a billion deaths if used by malicious actors.
What is the stance of some AI CEOs on regulating AI development pace?
The CEOs of Anthropic and OpenAI agreed on the need to slow down AI development, calling for federal and global cooperation and international standards.
How have US lawmakers responded to calls for AI regulation?
US House Speaker Mike Johnson declined to bring lawmakers back to regulate AI, deferring to the AI companies and citing concerns about rushing legislation.
Article produced with the assistance of artificial intelligence and reviewed by the editorial team.