The US offered China a hotline for AI incidents
While Trump publicly rejects AI guardrails, the US has proposed to China a notification mechanism for national security AI incidents.
The proposal
According to a Fortune report published on September 24, the US has proposed to China a notification mechanism between the two countries for national security related AI incidents. The talks span Treasury Secretary Scott Bessent's meeting with Chinese Vice Premier He Lifeng in New York earlier in the week and Xi Jinping's White House visit on September 23. The idea resembles the global alert systems Anthropic CEO Dario Amodei has called for.
A different public message
In public the administration spoke sharply against international oversight in the same days. White House science and technology adviser Michael Kratsios told the UN Security Council the US totally rejects efforts by international bodies to assert centralised control. At the White House, Xi said both countries have the capability and the responsibility to develop and manage AI.
Background
China launched the World Artificial Intelligence Cooperation Organization with 29 countries in July. Because the two countries are direct AI rivals, binding limits are not expected. An incident channel, much like crisis hotlines of the nuclear era, is seen as a narrow step meant to stop misunderstandings from escalating without halting the competition.