Breaking

U.S. Opens Door to AI Risk Talks With China Ahead of Trump-Xi Summit

September 16, 2026 at 13:44 EDT

Policy & Regulation
Foundation Models
AI Agents

The U.S. is open to discussing shared artificial-intelligence risks with China in multi-hour New York meetings this weekend, putting frontier-model safety on the agenda days before a Sept. 24 Trump-Xi summit. Treasury Secretary Scott Bessent's expected talks with Vice Premier He Lifeng would cover both open- and closed-weight models as Washington tries to avoid a split between U.S. and Chinese AI systems.

Sept. 16, 2026 · U.S.-China AI talks

AI risk talks move onto the agenda

U.S. officials are open to discussions with China on shared AI risks, system bifurcation and open- versus closed-weight models before a Sept. 24 leaders' summit.

New York

Multi-hour talks on AI, trade, rare earths and economic issues

Sept. 24

Trump-Xi summit where AI is expected to be part of the broader agenda

2 types

Open-weight and closed-weight models are both expected to be discussed

The diplomatic path

Weekend talksAI and economic agenda

→

Working levelPossible Monday follow-up

→

Leaders' summitSept. 24 milestone

Cited risk window: 6 to 12 months

A leading safety warning frames some agentic AI risks as a near-term window, shown here to scale.

6 months

12 months

Case for coordination

Shared standards, embedded evaluators and pre-release testing could slow the most hazardous frontier work without stopping all AI progress.

Key objection

Critics question independence, concentration of rulemaking power and whether rivals can cooperate while competing over chips, models and standards.

Continue reading

The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.