U.S.–China talks reportedly produce AI safety-notification proposal
A September 22 post describes an AI dialogue focused on uncontrollable agents, cyber threats involving non-state actors and major safety failures, with advanced AI-chip export controls kept separate.
TLDR
U.S. Treasury Secretary Scott Bessent and Chinese Vice Premier He Lifeng met in New York on September 20–21, 2026, to establish a formal AI dialogue, according to a September 22 post. It says the next meeting is scheduled for Shenzhen around November 2026 and that advanced AI-chip export controls remain separate from the safety channel. Another post describes the talks as producing an AI safety-notification proposal and a process to identify tariff cuts on non-strategic goods, while critical minerals, Boeing orders and farm purchases remain unresolved.
Combined views
—
3 Sources, first seen ago
