US and China ignore AI slowdown calls despite catastrophe warnings

Industry leaders are urging a worldwide slowdown in artificial‑intelligence development, but the United States and China – the two nations that dominate the field – have shown no sign of heeding the call. As President Donald Trump and President Xi Jinping prepare to meet in Washington, D.C. for their third summit this year, both sides have reiterated their commitment to press ahead with AI research rather than curb it.

The plea for a brake comes amid growing warnings from executives and researchers that unchecked AI could pose catastrophic, even existential, risks to humanity. Experts argue that because the U.S. and China host nearly all frontier labs and the majority of models downloaded worldwide, any safety framework that covers only one side would address merely half the problem. Yet mutual mistrust and divergent views of what threatens security are hampering cooperation. Trump has dismissed AI safety concerns as a “conspiracy,” insisting the United States must stay ahead of China in the technology and rejecting any “globalist scheme” to regulate it. China, while rolling out the third version of its AI security framework and warning of “hostile forces” using AI to spread political rumors, also rebuffs slowdown proposals, labeling Silicon Valley’s alarm as fear‑mongering and tying safety to broader concerns about cybersecurity, party authority and market access.

Analysts say the stalemate reflects deeper strategic competition. A senior fellow at the Asia Society Policy Institute noted that a bilateral safety regime is a “precondition” for any meaningful risk mitigation, but the two powers remain locked in a race for dominance. Chinese security officials have highlighted political, institutional and ideological threats, while U.S. officials prioritize maintaining a technological edge. With the summit unlikely to produce a coordinated slowdown, observers expect the rivalry to continue driving rapid AI advancement, raising the stakes for international dialogue on safety standards. The next step may be a series of bilateral technical talks, but without trust or a shared definition of risk, the prospect of a global safety pact remains distant.