The proposed AI incident channel between the United States and China now has a tentative timetable, but not yet an agreement. Trump administration officials told Wired that formalizing the system could take several weeks, even as they expressed optimism that it might operate before the end of 2026.
That reporting adds concrete obstacles to the initial announcement of an AI notification mechanism. The two governments still have to decide what qualifies as a national-security incident, when an alert becomes mandatory and how obligations would be enforced.
China Has Not Formally Signed On
Treasury Secretary Scott Bessent first announced the proposal with Chinese vice premier He Lifeng. The concept resembles a Cold War hotline: each country would have a route for notifying the other when an AI event threatens national security. Chinese support, however, remains noncommittal according to US officials cited by Wired.
The administration is targeting a functional framework that President Donald Trump can approve by the time Bessent and Lifeng meet again in Shenzhen in about two months. Officials have discussed benchmarks that would determine when an incident crosses the national-security threshold.
The talks are unfolding as Trump and Chinese president Xi Jinping prepare for a White House state dinner. OpenAI and Nvidia executives are expected to attend, underscoring how closely the diplomatic discussion is tied to the companies supplying models and computing infrastructure.
Those benchmarks are not public, and no final timetable has been agreed. The year-end objective is an expectation from US officials rather than a joint commitment. The difference matters because a hotline cannot provide predictable crisis communication unless both governments accept the trigger conditions and the information each side must disclose.
Enforcement Is Entangled With the Chip Dispute
US officials are considering measures designed to prevent the channel from becoming a one-way system in which Washington sends warnings without reciprocal information from Beijing. Wired reports that the framework may specify criteria for alerts and consequences intended to encourage compliance.
The negotiation sits beside a broader trade and technology dispute. Trump is expected to extend a trade truce and has been reluctant to end limited licenses allowing Nvidia to sell H200 chips to China. Supporters of those sales argue that restrictions encourage Chinese companies to develop domestic alternatives, while Huawei is already reported to have commercial AI chips.
The proposed channel echoes an earlier US-China financial dialogue. Wired notes that former Treasury Secretary Hank Paulson credited that relationship with helping avert a large Chinese sale of US-backed debt during the 2008 financial crisis. The comparison is an argument for communication, not evidence that an AI mechanism would produce the same result.
Those issues do not establish whether an incident channel will work, but they complicate enforcement. Each government may see safety information through the same competitive lens that shapes access to computing hardware. A technically sound notification procedure still requires political confidence that sharing a warning will not create a strategic disadvantage.
Leadership Decisions Could Add Another Delay
Some administration officials believe progress may depend on Trump’s choice of a new AI coordinator. The proposed “AI czar” would lead interagency work through an envisioned AI Force. Bessent, Commerce Secretary Howard Lutnick, Michael Kratsios and Sean Cairncross have all been discussed for the role, according to Wired’s sources.
The article does not say that an appointment is legally required before an agreement can be completed. It reports an internal concern that unresolved leadership could slow the process. That uncertainty joins the larger list of unfinished details: Chinese consent, thresholds, enforcement and White House approval.
David Sacks, who previously held the AI role as a special government employee, is not expected to return, according to the sources. It is also unresolved whether the next coordinator would keep another government position or serve in the AI role full time.
The new facts make the hotline more tangible but also less immediate. Washington and Beijing have moved beyond a vague expression of interest, yet the mechanism remains a negotiating project. Its value will depend on rules precise enough to function during a crisis and flexible enough for two strategic rivals to accept.
