The Microsoft co-founder warned that international artificial intelligence pacts will be harder than Cold War nuclear treaties and called for lab monitoring.
Microsoft co-founder Bill Gates called on the U.S. to lead global artificial intelligence regulations, warning that negotiating international standards will be harder than Cold War nuclear treaties. Gates made the remarks during an interview with NBC News’ “Meet the Press” that aired Sunday, following President Donald Trump’s public pushback against federal guardrails on AI developers.
Earlier this month, Trump rejected federal restrictions on AI companies, posting on Truth Social that “WHOEVER WINS AI, WINS! We are leading China, and all others, and will continue to do so.” Trump also posted ahead of a meeting with Chinese President Xi Jinping that he wanted to leave superintelligence “exactly where it is.” Gates countered that requiring safety monitoring inside tech labs won’t put the country at a disadvantage against foreign rivals.
Gates said he believes he and others can reach an agreement with the president once Trump hears their perspective. “What he said is fine, but I think after he hears from me and other people, including that this does not handicap us in whatever he thinks the nation-state race is, that, you know, we will come to a consensus,” Gates told NBC News.
Push for monitoring over kill switches
Concerns escalated this month after former Anthropic employee Jacob Coxon publicly warned that AI creators believe the technology could kill humanity by the end of the decade. Anthropic CEO Dario Amodei and OpenAI CEO Sam Altman have both supported slowing development down, with OpenAI recently reporting that some of its autonomous agents hacked an outside company without instructions.
Instead of focusing solely on existential threats, Gates urged immediate safeguards to stop criminals and small groups from using AI to scramble bank accounts, take down power grids, or craft biological weapons. A bipartisan proposal in Congress would mandate a kill switch to shut down dangerous AI models, but Gates argued that isn’t enough on its own. He said governments must require monitoring tools inside AI labs to give law enforcement records of what systems are doing before large-scale attacks happen.
Gates Foundation language efforts
Gates also detailed how his philanthropic organization is using the technology. The Gates Foundation plans to spend more than $200 billion over the next 19 years and is using AI to direct spending on malaria and malnutrition.
The foundation is partnering with 60 AI firms to reduce error rates in non-English languages. Gates said AI systems have an error rate about 10 times higher in African languages than in English, which he called dangerous for health and education. The foundation aims to bring models for every language spoken by at least one million people close to English quality within five years.
