Illustration for: US Pitches China An AI Risk Alert System

US Pitches China An AI Risk Alert System

Treasury Secretary Scott Bessent proposed a US-China AI safety notification mechanism covering national-security-level AI incidents during talks in New York, laying groundwork for this week's Trump-Xi summit.

By the Numbers

Sept 20, New York
Talks held
National-security AI incidents
Scope
Off agenda
Chip export controls
This week, Washington
Trump-Xi summit
TC
By the Markets Desk
Edited by Trace Cohen · Early-stage VC & angel · Founder, New York Venture Partners
2 min read
ShareXLinkedInEmail

THE RUNDOWN

1

The proposal specifically covers AI-related incidents that rise to a national-security level, not general AI safety cooperation -- a narrower, more targeted scope than a broad bilateral AI safety treaty would carry.

2

US Trade Representative Jamieson Greer explicitly said export controls on advanced AI chips and semiconductor manufacturing equipment were not on the agenda for these particular talks -- meaning the notification mechanism is a separate, narrower track from the chip-export fight that has otherwise defined US-China AI policy.

3

Bessent framed the goal as moving 'from opaque to more transparency between the number one and the number two AI powers in the world' -- language that treats disclosure itself, not any specific enforcement mechanism, as the primary goal of this proposal.

4

The talks were explicitly positioned as groundwork for a 'successful summit' between Trump and Xi the following week in Washington -- meaning this proposal's actual substance may depend heavily on what, if anything, gets formally agreed at that higher-level meeting.

TC

The VC Read · Trace's Take

Trace Cohen

Keeping chip export controls off the table is what makes this proposal achievable at all -- the two governments carved out the one issue that would sink the talks and left the notification mechanism as the part both sides can agree costs them little. The thing to actually watch isn't Bessent's framing, it's whether the Trump-Xi summit gives this any formal weight; a proposal from preparatory talks that never gets mentioned at the main event is a talking point, not a policy.

Analysis

US Treasury Secretary Scott Bessent proposed a new AI safety notification mechanism with China during talks with Chinese Vice Premier He Lifeng in New York on September 20, aimed at setting up a system for the two countries to alert each other to AI-related incidents that reach a national-security threshold, according to CNN and NBC News.

A Narrow Mechanism, Not A Broad Treaty

The proposal is specifically scoped to AI incidents that rise to a national-security level, rather than a general AI safety cooperation framework -- a narrower ambition than a comprehensive bilateral AI treaty would represent, but also a more achievable one given the current state of US-China relations. Bessent described the goal directly: "moving from opaque to more transparency between the number one and the number two AI powers in the world is very important."

Separating the two tracks lets both sides make progress on incident transparency without reopening the far more contentious chip-access fight.

What's Explicitly Off The Table

US Trade Representative Jamieson Greer said export controls on sophisticated AI chips and semiconductor manufacturing equipment were not part of these particular discussions -- keeping the notification-mechanism talks entirely separate from the chip-export restrictions that have otherwise defined the US-China AI relationship all year, including the export-control pressure pushing DeepSeek toward Huawei's Ascend chips covered elsewhere in this issue. Separating the two tracks lets both sides make progress on incident transparency without reopening the far more contentious chip-access fight.

Setting Up A Bigger Meeting

Greer described Sunday's talks as groundwork for a "successful summit" between Trump and Xi the following week in Washington -- meaning the notification mechanism as currently described is preparatory, not a finished or signed agreement. Whether it survives contact with the actual Trump-Xi summit intact, gets expanded, or quietly narrows further is an open question that depends on dynamics well above Bessent and Greer's negotiating table.

Why This Matters For AI Investors

For any US company or investor with China-linked AI exposure -- whether through supply chains, joint ventures, or portfolio companies operating across both markets -- a functioning bilateral notification channel would be a genuinely useful de-escalation tool, reducing the chance that a serious AI incident in one country triggers an uncoordinated, worst-case response from the other. But a proposal discussed in preparatory talks is not the same as an operating system, and the history of US-China technology negotiations this year suggests the gap between proposal and implementation can be wide.

What To Watch

Whether this notification mechanism gets any formal mention or commitment at the Trump-Xi summit later this week is the immediate marker of whether it has real momentum behind it, or whether it remains a talking point from one round of preparatory diplomacy that doesn't survive to the main event.

ShareXLinkedInEmail

Key Sources

2 sources
SourceCNN

Reported by CNN · Analysis by Value Add Pulse.

← Back to Pulse

THE WIRE in your inbox— Tech, startup & VC news with Trace's take. Free, no spam.