US Proposes AI Notification Mechanism in Talks With China
publishers 6articles 6first reported 23 Sep, 21:00 UTCdeveloping since 24 Sep · 5 editions
The United States has proposed an artificial intelligence notification mechanism with China that would allow the two governments to alert each other when AI-related incidents threaten national security, US Treasury Secretary Scott Bessent said ahead of a meeting between President Donald Trump and Chinese President Xi Jinping in Washington.
The mechanism has been likened to a hotline, but an agreement is not yet complete. Wired reported that Trump administration officials expect negotiations to take several more weeks and hope to establish a functional framework before Bessent’s next meeting with Chinese Vice Premier He Lifeng, planned for two months from now in Shenzhen. China has not explicitly agreed to the proposal, according to the report.
Outstanding issues include defining which incidents would require notification and creating measures to ensure that both countries participate. US officials have discussed criteria governing alerts and possible enforcement provisions, Wired reported.
Xi said Thursday that China and the United States, as leading countries in AI, have a responsibility to develop and manage the technology for beneficial purposes and ensure it remains under human control, according to The Hill.
The discussions come as the countries compete in AI development. Al Jazeera reported that the United States retains a large advantage in computing capacity and infrastructure, while Chinese companies are narrowing the gap in advanced models and gaining users partly through lower prices.
HOW THIS STORY WAS MADE
Written from 5 articles, 3 with the publisher's own text; 5 independent newsrooms once syndicated copies count as one; this version written within the hour the record first saw the story; the editor kept it as written.
- Publishers
- 6
- Source articles
- 6
- Given to the writer
- 5, 3 with the article's own text
- Independent newsrooms
- 5
- This version written
- within the hour the record first saw the story
- Second model (editor)
- kept as written
- Publication gate
- passed
- Human review
- none
Written by a language model from the sources above, then checked by a second model that may only cut, attribute or correct. How it works →