On 24 September, the two capitals finally talk about artificial intelligence. The agenda is broad, the ambition deliberately modest: to set definitions, not a framework.
The first bilateral dialogue on artificial intelligence between the United States and China was held on September 24, on the sidelines of the meeting between Donald Trump and Xi Jinping. It is a direct result of the May 2026 summit in Beijing. On the American side, the delegation is led by Treasury Secretary Scott Bessent.
The agenda covers joint monitoring of model-driven cyberattacks, Chinese access to U.S. border models, U.S. restrictions on Chinese open-weight models, and whether Beijing will regulate open-source AI. The common goal is to prevent the most capable systems from falling into the hands of non-state actors.
Expectations remain low. According to CSIS's analysis, this first round aims to establish shared definitions and working channels — including incident reporting and a minimum set of security practices — rather than a comprehensive framework. Recent failures to contain autonomous agents and industry calls to slow down the race to the border were expected to weigh on this.
Read the full article on the original source
Go to the source →