Bessent and He discuss US-China AI communications in New York
On September 20, 2026, US Treasury Secretary Scott Bessent and Chinese Vice Premier He Lifeng met in New York for talks that included artificial intelligence, trade and agenda items for a coming leaders’ summit. The sides agreed to formalize an AI dialogue, but accounts differed on whether they had also agreed to establish an incident hotline or were still exploring one.
Key points
- The meeting produced agreement to formalize government-to-government dialogue on AI.
Evidence · 3 source statements
The two sides have already agreed to establish government-to-government dialogue on AI.
US Treasury Secretary Scott Bessent said after the New York meeting that the two sides had agreed to formalize an AI dialogue, explore an "incident line" for significant AI…
The sides also agreed to launch an artificial intelligence dialogue to prevent escalation.
- The status of a separate channel for AI safety incidents was described variously as proposed, under exploration or agreed.
Evidence · 4 source statements
On Sunday, Scott Bessent, the US treasury secretary, concluded talks with the Chinese vice-premier He Lifeng by proposing an AI safety mechanism to be established between…
US Treasury Secretary Scott Bessent said after the New York meeting that the two sides had agreed to formalize an AI dialogue, explore an "incident line" for significant AI…
- Bessent identified potential discussion topics including uncontrollable AI agents, cyberattacks by non-state actors and AI-assisted biological weapons.
Evidence · 1 source statement
Who's involved
Scott Bessent
The US Treasury secretary took part in the talks and described the proposed AI cooperation.
Bessent said greater transparency between the two leading AI powers was important and said the sides had agreed to formalize dialogue while exploring an incident line.
What they said · 2 source statements
We think that, just like with any cross-border activity, that moving from opaque to more transparency between the number one and the number two AI powers in the world is very…
US Treasury Secretary Scott Bessent said after the New York meeting that the two sides had agreed to formalize an AI dialogue, explore an "incident line" for significant AI…
Where accounts differ
Whether the sides had agreed to establish an AI incident hotline or notification mechanism
- Some accounts described the mechanism as proposed or still being explored.
Evidence · 3 source statements
On Sunday, Scott Bessent, the US treasury secretary, concluded talks with the Chinese vice-premier He Lifeng by proposing an AI safety mechanism to be established between…
US Treasury Secretary Scott Bessent said after the New York meeting that the two sides had agreed to formalize an AI dialogue, explore an "incident line" for significant AI…
The secretary proposed an AI "notification mechanism" between the U.S. and China that would allow the two countries to alert each other about AI-related safety incidents that…
- Other accounts said the sides agreed to a mutual notification mechanism or hotline.
Evidence · 2 source statements
The accounts differ on whether agreement covered only the formal AI dialogue or also the incident communication channel.
What was reported
- He Lifeng and Scott Bessent met in New York on September 20, 2026; they posed for a photo and shook hands before consultations.
Reported independently by 2: CGTN, Reuters
Evidence · 3 source statements
Chinese Vice Premier He Lifeng, also a member of the Political Bureau of the Communist Party of China Central Committee, and US Treasury Secretary Scott Bessent posed for a photo…
Chinese Vice Premier He Lifeng (R), also a member of the Political Bureau of the Communist Party of China Central Committee, shakes hands with US Treasury Secretary Scott Bessent…
- The talks included trade and artificial intelligence, and the two officials discussed preliminary agenda items for a US-China leaders’ summit.
Reported independently by 2: BBC News, Reuters
Evidence · 2 source statements
very successful
- CGTN reported that China and the United States had agreed to establish government-to-government dialogue on AI; TASS also reported agreement to launch an AI dialogue aimed at preventing escalation.
Reported independently by 2: CGTN, TASS
Evidence · 2 source statements
The two sides have already agreed to establish government-to-government dialogue on AI.
The sides also agreed to launch an artificial intelligence dialogue to prevent escalation.
- Bessent said the two sides agreed to formalize an AI dialogue, explore an incident line for significant AI safety events and continue discussions in Shenzhen in about two months.
One source: CGTN
Evidence · 1 source statement
US Treasury Secretary Scott Bessent said after the New York meeting that the two sides had agreed to formalize an AI dialogue, explore an "incident line" for significant AI…
- Other accounts described a mutual notification mechanism or hotline for serious AI incidents, including incidents that could threaten national security.
Reported independently by 4: CNBC, El País, Fox News, The Hankyoreh
Evidence · 4 source statements
The secretary proposed an AI "notification mechanism" between the U.S. and China that would allow the two countries to alert each other about AI-related safety incidents that…
- US Trade Representative Jamieson Greer also attended the Sunday meeting.
One source: South China Morning Post
Evidence · 1 source statement
Sunday’s meeting was also attended by US Trade Representative Jamieson Greer.
Still unclear
- The hotline’s scope, operators and procedures were not specified.
Not heard from
- The Chinese government and He Lifeng — They were directly involved, but the claims contain no statement from them explaining their position on the AI dialogue or incident channel.