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

Meeting in New York

US Treasury Secretary Scott Bessent reported that he held talks with Chinese Vice Premier He Lifeng in New York, describing the discussions as successful. The agenda included the creation of a new notification mechanism intended to alert both governments to AI incidents that might threaten national security. Bessent emphasized the importance of moving from opacity to greater transparency between the two leading AI powers.

Context of the Summit

The meeting took place as President Donald Trump and Chinese President Xi Jinping prepare for a summit in Washington later in the week. The summit follows previous high-level exchanges on AI, including talks in Beijing earlier in the year that linked AI cooperation to trade agreements involving Boeing aircraft and US agricultural products. A truce in the broader tariff dispute between the United States and China is set to expire on 10 November.

Statements from Both Sides

Chinese state news agency Xinhua said the two sides engaged in candid, in-depth and constructive exchanges on key economic and trade issues, noting that AI was part of the discussion. Bessent later posted on X that the talks help lay groundwork for President Trump to advance American economic interests. The BBC has sought comment from the Chinese embassy, but no response was received at the time of publication.

Unverified Details

Specifics of the proposed notification mechanism, including its operational framework and any timelines, were not disclosed. It remains unclear how the mechanism would be integrated into existing diplomatic or security channels, and whether any formal agreement will be reached before the summit.