US-China AI Alert System Targets No. 1 and No. 2 AI Powers' Cyber Risk

For cybersecurity teams, an AI incident between the US and China is not just a diplomatic abstraction—it is a national security escalation vector that could arrive as a compromised model, an AI-driven intrusion, or a misattributed autonomous system failure. The proposed notification mechanism would create a channel to distinguish accident from attack before response decisions cascade.

On September 20, 2026, Treasury Secretary Scott Bessent proposed a bilateral “notification mechanism” for artificial intelligence-related incidents that could affect national security, during talks in New York with Chinese Vice Premier He Lifeng. The proposal, reported by Fox affiliates on September 21 and 22, comes ahead of President Donald Trump's White House meeting with Chinese President Xi Jinping on September 24. Bessent called the two countries “the No. 1 and the No. 2 AI powers in the world” and said moving “from opaque to more transparency” is important.

“

On September 20, 2026, Treasury Secretary Scott Bessent proposed a bilateral “notification mechanism” for artificial intelligence-related incidents that could affect national security, during talks in New York with Chinese Vice Premier He Lifeng.

The mechanism is a confidence-building measure, similar in logic to nuclear risk reduction or cyber hotlines, but applied to AI. No technical architecture, trigger thresholds, or verification process has been disclosed, and China has not formally accepted it. Still, the proposal signals that both governments recognize an accidental AI failure, misattributed autonomous action, or AI-enabled cyber incident could escalate quickly. For Washington, the goal is a channel to distinguish accident from deliberate attack; for Beijing, it may raise concerns about exposing sensitive capabilities.

The proposal sits within the Trump administration's broader AI acceleration policy. Trump has resisted calls to slow AI development and has announced an “AI Force” modeled on Space Force plus an “AI Czar” to oversee national AI development. That makes the notification idea distinctive: it is not a moratorium or an export-control agreement, but a safety communication layer that could coexist with intense strategic competition.

The AI push is advancing alongside trade diplomacy. Bessent and US Trade Representative Jamieson Greer announced that groundwork was in place for Xi's visit and that the two countries agreed to move forward with a “Board of Trade” first discussed at the May 2026 Beijing summit. Greer said negotiators are still working out which “nonsensitive” items could receive lower tariffs, potentially including consumer and agricultural goods, energy products, and medicinal devices. This suggests the AI mechanism is part of a larger framework of selective de-escalation while high-end technology competition continues.

For industry, the proposal could eventually influence how AI incidents are classified, disclosed, and managed across borders. AI developers, cloud providers, cybersecurity vendors, and semiconductor firms may face new compliance expectations if governments define national-security AI incidents and require reporting. On the trade side, lower tariffs on nonsensitive categories would help consumer technology, energy equipment, and medical device supply chains, though advanced AI hardware and sensitive technologies are likely to remain controlled.