The Pulse
U.S. and China Agree to a Channel for SI Incidents
The White House says Donald Trump and Xi Jinping agreed to create a bilateral channel for “super intelligence” incidents. A Xinhua news report published on China.gov.cn describes the leaders’ discussion of AI cooperation but does not mentio
AI.info Team ·
The White House says Donald Trump and Xi Jinping agreed to create a bilateral channel for “super intelligence” incidents. China’s public account of the same summit emphasizes continued AI dialogue and cooperation, but does not mention the incident channel—leaving a gap between Washington’s specific announcement and Beijing’s description of the talks.
The agreement was announced in a White House fact sheet dated September 25, after the presidents met at the White House on September 24. The document says the two countries also established a U.S.-China Super Intelligence Dialogue to discuss risks and benefits, with another exchange scheduled by November 2026.
Washington names an incident channel; Beijing stresses dialogue
The U.S. fact sheet presents the incident channel as a joint commitment: “The United States and China also agreed to establish a bilateral communication channel for SI incidents.” It gives no further description of what counts as an incident, which officials would use the channel, or how notifications would be handled.
A Xinhua news report published on China.gov.cn about the leaders’ September 24 talks takes a broader approach. It says the two presidents discussed AI and quotes Xi calling for continued exchanges on risks and benefits and cooperation to prevent misuse. The report does not mention an incident channel or the newly named Super Intelligence Dialogue.
“The two sides can continue to engage in dialogue on AI, exchange views on its risks and benefits, and work together to prevent the misuse and abuse of AI.”
Xi Jinping, Chinese president, quoted by Xinhua
The difference in the public accounts does not establish that Beijing rejects the channel.
“Super intelligence” is the agreed label
The White House says the leaders agreed to use “super intelligence” rather than “artificial intelligence” for the relevant emerging technologies. Its fact sheet places the incident channel alongside the new dialogue on the risks and benefits of those technologies, but does not define the term or state that every AI-related failure would qualify for notification.
That leaves the scope of the channel unclear. The announcement does not say whether it covers accidents, malicious use, failures with cross-border effects, or events that governments judge to pose a security threat. Nor does it identify an activation threshold, a response timeline, or the agencies that would manage communications.
The November exchange is the next stated milestone
The White House says the next U.S.-China Super Intelligence Dialogue exchange will occur by November 2026. It does not say whether that meeting will settle operating procedures for the channel, or whether the mechanism will be active before then.
For now, the concrete public commitment is narrow: the White House says both governments agreed to establish a bilateral channel for SI incidents, while the Xinhua report published on China.gov.cn describes continued AI dialogue without referring to that mechanism. The announced November exchange is the next date the fact sheet assigns to the new dialogue.