Full Breakdown
U.S.–China AI Safety Talks Ahead of Trump-Xi Summit
By Drooid · · How we work
Core Event: Proposal of an AI-Incident Notification Mechanism
U.S. Treasury Secretary Scott Bessent announced that, after day-long talks in New York with Chinese Vice Premier He Lifeng, the United States has put forward a new “U.S.–China AI dialogue” that would include a mechanism for notifying each side of AI-related incidents that could affect national security. “We think that, just like with any cross-border activity, moving from opaque to more transparency between the number one and the number two AI powers in the world is very important,” — Scott Bessent.
Background & Context
The talks come as President Donald Trump and President Xi Jinping prepare to meet in Washington later this week, their third summit of the year. “The two countries host nearly all frontier labs and most of the models people download,” — Alvin Wang Graylin. Both sides have so far rejected broader slowdown proposals, with the United States emphasizing the need to stay ahead and China framing safety concerns within cybersecurity and party-authority considerations.
Official Statements & Responses
- Scott Bessent said transparency on AI incidents is “very important” and described the New York talks as “very successful,” focusing on “common goals and common threats.”
- Jamieson Greer, U.S. Trade Representative, clarified that export controls on advanced AI chips were not on the agenda for the AI mechanism, noting the broader trade truce expires on November 10.
- President Trump dismissed AI-related existential warnings as a “hoax,” asserting that “whoever wins in AI, wins,” and warned the United States must not allow China to pull ahead.
- Chinese officials have not detailed their response; state media reported “frank, in-depth, and constructive exchanges” on economic and trade matters, including AI, but offered no comment on the notification system.
Criticism & Opposition
Industry leaders Anthropic CEO Dario Amodei and OpenAI CEO Sam Altman have called for a slowdown and greater regulation. David Krueger, assistant professor at the University of Montreal, argued that “international coordination is absolutely critical. We need more than safeguards.” In contrast, Nvidia CEO Jensen Huang and Meta CEO Mark Zuckerberg downplay external oversight, emphasizing AI’s transformative benefits.
Conflicting Reports & Gaps
- Scope of Cooperation: U.S. officials envision a modest notification channel, while some researchers argue only an “indefinite international moratorium” would address existential risks.
- Export Controls: Greer said chip export controls were excluded from the AI talks, yet analysts note the United States continues to restrict China’s access to advanced AI chips, a point China contests.
- Chinese Position: State media highlighted constructive economic dialogue but did not confirm acceptance of the notification proposal, leaving China’s stance ambiguous.
What’s Next
The AI safety dialogue will be discussed at the Trump-Xi summit later this week in Washington. Both sides have indicated a willingness to meet again on AI issues, suggesting the notification mechanism could become a standing element of bilateral engagement. The broader U.S.–China trade truce, underpinning many economic discussions, expires on November 10, adding urgency to any forthcoming agreements.