US Proposes AI Incident Alerts to China Before Trump-Xi Summit

The United States and China are exploring a channel to notify each other of artificial intelligence incidents that could affect national security, a step that would mark the first structured bilateral effort to manage risks from the technology even as the two countries compete to lead its development.

Treasury Secretary Scott Bessent described the proposal after meetings in New York on Sunday with Chinese Vice Premier He Lifeng. The plan would create a US-China AI Dialogue that includes an alert mechanism and follow-up talks on shared goals and threats. Bessent said the two sides agreed to meet again on the issue and that the framework would be presented to President Donald Trump and Chinese leader Xi Jinping at their summit in Washington later this week.

Bessent stated that moving “from opaque to more transparency between the number one and the number two AI powers in the world is very important.” The discussions built on earlier exchanges during Trump’s May visit to Beijing. US Trade Representative Jamieson Greer said the AI talks did not cover existing US export controls on advanced chips and semiconductor equipment.

The same meetings also advanced work on a new Board of Trade and identified categories of non-sensitive goods that could receive different tariff treatment in future disputes. Greer said those categories could include certain consumer goods from China and US exports of energy, agricultural products, and medical devices. A fragile trade truce between the two countries is set to expire November 10.

China’s state news agency Xinhua reported that the talks covered AI along with economic and trade issues but did not mention the notification proposal. Chinese officials did not comment on the US suggestion when leaving the meetings. Experts have noted that mutual distrust makes a major AI agreement at the summit unlikely, and China has viewed calls to slow AI development as aimed at limiting its progress.

Details on specific incidents that would trigger notifications, technical definitions of national security threats, or any prior bilateral AI safety talks were not released by either side.