Bessent Unveils US-China 'Superintelligence' AI Dialogue

The US and China are launching a bilateral dialogue on "superintelligence," with the first meeting set for Shenzhen on the sidelines of the Asia-Pacific Economic Cooperation summit on November 18-19, Treasury Secretary Scott Bessent said in prepared remarks to the National Committee on US-China Relations, as reported by Reuters.

Using the shorthand "SI," Bessent said the two governments will share approaches to assessing advanced AI models before they are released and to keeping human judgement in control. They are also building a dedicated line of communication through which either nation can warn the other about grave incidents tied to its superintelligence models.

The two powers should "harbor no illusions" about their differences on AI even as they engage to contain its risks, Bessent said, with the administration's framing holding at "de-risk, not decouple."

He stopped short of making any commitment to limit advancement of the technology, as Washington and Beijing remain locked in an intense race for AI supremacy. Invoking the first US trade voyage to China in 1784, Bessent said the relationship would be sustained by recognizing "where common purpose must prevail over competing interests."

The channel lands inside a dense run of China-AI policy coverage on our tracker, including an Axios account of Bessent confronting Lutnick over a Megaspeed entity-list delay.