US and China discuss AI safety plan ahead of Trump-Xi summit

Top US and Chinese officials have discussed creating a new “notification mechanism” for AI incidents that could affect national security, Treasury Secretary Scott Bessent told reporters on Sunday.

Bessent made the comments after what he called “successful” talks with Chinese Vice Premier He Lifeng in New York.

The meeting came as US President Donald Trump and his Chinese counterpart Xi Jinping are due to hold a summit in Washington later this week.

AI has come under intense scrutiny in recent days after researchers warned of the potential risks posed by the technology and some of the industry’s most high profile figures called for developers to take a more cautious approach.

“We think that, just like with any cross-border activity, that moving from opaque to more transparency between the number one and the number two AI powers in the world is very important,” Bessent told reporters after the talks.

He said the meeting also covered other topics, including plans to “operationalise” a process to identify potential tariff cuts on goods called the Board of Trade.

A truce in the tariffs war between the world’s two biggest economies is due to expire on 10 November.

The BBC has contacted the Chinese embassy in the US for comment.

Chinese state news agency Xinhua said the two sides had “candid, in-depth and constructive exchanges on key economic and trade issues”, noting that they had discussed AI.

Bessent said in a post on X on Sunday that the talks “help lay the groundwork for President Trump to advance America’s economic interests and deliver results for the American people.”

The US and China are locked in a race to dominate the AI and technology industries.

In recent days Washington has faced calls to slow down the development of AI over fears abouts its potential negative impact on humanity.

Trump has largely dismissed those concerns, arguing that there are sufficient policies around AI and that the US cannot lose its edge to China.

He said earlier this month “whoever wins in AI, wins.”