The US and China have agreed to a joint AI dialogue, with a focus on a notification mechanism for security-related incidents, according to reports from the Financial Times and The Information. US Treasury Secretary Scott Bessent announced the move after talks with Chinese Vice Premier He Lifeng in New York. The mechanism is designed to raise the alarm when AI incidents impact national security, aiming to move from opacity to greater transparency between the two nations. Both sides plan another meeting to work out the details of the reporting system.

US Trade Representative Jamieson Greer clarified that export restrictions on advanced AI chips were not part of the discussions. The announcement comes as President Trump prepares to host Xi Jinping at Joint Base Andrews in Washington. The summit is the first on US soil since 2017 and follows a trade truce reached in Busan, South Korea. Despite industry warnings about losing control of AI, Trump has rejected any deal to slow AI research, prioritising the tech race with China.