US and China outline AI safety notification plan ahead of Trump-Xi summit

Treasury Secretary Scott Bessent emerged from talks in New York on Sunday with news that the United States and China had begun discussing a formal system for notifying each other when artificial intelligence incidents threaten national security. The meeting, held between Bessent and Chinese Vice Premier He Lifeng, marked a rare moment of cooperation between the world's two dominant AI powers at a time when both nations are locked in an intense competition to control the technology's future.

Bessent described the talks as successful and framed the proposed notification mechanism as a straightforward matter of transparency. "We think that, just like with any cross-border activity, that moving from opaque to more transparency between the number one and the number two AI powers in the world is very important," he told reporters. The conversation extended beyond AI safety to include discussions about tariff policy, including plans to operationalize a process for identifying potential tariff cuts through what the administration calls the Board of Trade. A temporary truce in the tariff war between the two countries is set to expire on November 10.

The timing of these talks matters. Bessent and He Lifeng met just days before President Donald Trump and Chinese leader Xi Jinping are scheduled to hold a summit in Washington later in the week. The meeting in New York served as preliminary groundwork, with Bessent stating afterward that the discussions would "help lay the groundwork for President Trump to advance America's economic interests and deliver results for the American people." Chinese state news agency Xinhua characterized the exchange as featuring "candid, in-depth and constructive exchanges on key economic and trade issues," confirming that AI was on the agenda.

The proposal for an AI notification system arrives at a moment when the technology has drawn intense scrutiny from researchers and prominent industry figures. In recent days, warnings about artificial intelligence's potential risks have circulated widely, with some of the field's most visible voices calling for developers to adopt more cautious approaches to deployment and advancement. These concerns have created pressure on governments to establish guardrails around AI development.

Trump has largely resisted that pressure. Earlier this month, he stated plainly that "whoever wins in AI, wins," signaling his view that the United States cannot afford to slow its technological progress. He has argued that existing policies adequately address AI concerns and that losing ground to China in this domain would be strategically catastrophic. His administration's focus remains on maintaining American dominance in the field rather than on restricting development in the name of safety.

The notification mechanism under discussion would represent a modest but meaningful shift in how the two nations approach AI governance. Rather than operating in isolation, each country would alert the other to incidents that could cross borders or affect the other's security. It is a framework borrowed from other domains of international relations—the kind of transparency measure that typically emerges when two competitors recognize that some shared rules benefit both sides.

Whether this cooperation will survive the Trump-Xi summit and become a lasting feature of US-China relations remains uncertain. The two nations remain fundamentally at odds over technology, trade, and geopolitical influence. The notification mechanism could become a cornerstone of AI governance between them, or it could be overshadowed by the competitive pressures that dominate the relationship. The summit this week will offer the first real test of whether both leaders view AI safety cooperation as a priority or as secondary to their broader strategic interests.