Washington is exploring a high-level artificial intelligence dialogue with Beijing, but deep mutual distrust could limit any meaningful outcomes
The Trump administration is weighing whether to hold an executive-level meeting on artificial intelligence during Chinese President Xi Jinping’s upcoming visit to Washington, a move that would place AI governance at the center of the world’s most consequential geopolitical relationship.
Xi’s state visit is scheduled for September 24, 2026, and the potential AI dialogue would build on conversations the two leaders held during their May 2026 summit in Beijing. That meeting produced discussion around bilateral AI “guard rails” but no formal framework or signed agreement.
What’s on the table
Treasury Secretary Scott Bessent is reportedly leading US efforts to weave AI safety conversations into the broader economic agenda surrounding Xi’s visit. The administration’s interest appears driven by a specific and growing concern: AI-driven cyberattacks.
One proposal under consideration involves collaboration between American and Chinese AI labs to share information aimed at preventing AI-related cyber threats. Progress on that front has been slow, with credibility concerns on both sides making even preliminary information-sharing arrangements difficult to negotiate.
By early September 2026, a White House official confirmed that no AI-specific meeting had been formally scheduled for mid-September, even as negotiations continued behind the scenes.
The May summit left a lot unfinished
The Beijing meeting in May set the stage for these discussions. Trump and Xi talked about establishing AI guardrails and addressed the thorny issue of Nvidia chip exports to China. No formal AI framework emerged from that summit.
OpenAI CEO Sam Altman and Anthropic CEO Dario Amodei have both advocated for safety agreements or pacing measures designed to reduce the risks posed by increasingly powerful AI systems.
