U.S. and China Agree to Create AI Incident Channel After Washington Summit

The U.S. and China agreed to establish a channel for AI-related incidents after a three-day summit in Washington. They also committed to keep discussing the technology’s risks and benefits. The channel promises a way to communicate about specific problems, but the public agreement is not yet a guide to how either government would raise one.

A dialogue and a separate incident channel

The White House calls the forum for regular discussion the U.S.-China Super Intelligence Dialogue. It says the two countries established it to exchange views on risks and benefits, with the next exchange due by November. The bilateral channel for incidents is a separate commitment. One creates time to discuss the technology; the other is meant to provide a route for communicating about an incident.

The wording differs across the summit accounts. The White House says the leaders agreed to use “super intelligence” rather than “artificial intelligence” for the applicable emerging technologies. In its description of China’s statement, the Associated Press uses artificial intelligence for the incident mechanism and says an AI-specific dialogue is scheduled for November. Both accounts describe continued talks and a channel for incidents.

Other channels from the summit

The leaders’ meeting also produced steps on military and trade communications. China’s Foreign Ministry said the governments agreed to sign a memorandum on strengthening crisis communications and preventing crises between their militaries. That is a commitment to sign, not confirmation that the document is already in force. Separately, the White House said the countries put their Board of Trade into operation during the summit week to work on selected bilateral trade issues.

AP assessed the summit as producing no major breakthroughs. Wang Zichen of the Beijing-based Center for China and Globalization offered a narrower measure of progress: he said the emphasis on working-level engagement and crisis management suggested an effort to make stability last beyond the leaders’ personal diplomacy. For the AI commitment, the next exchange offers officials a chance to do that work; whether it yields a usable incident process remains unresolved.