U.S. And China Vow Greater Cooperation On AI, Announce Creation Of Nuclear-Hotline-Style Communications Channel.

After weeks of growing calls by prominent figures in both tech and politics for more stringent AI regulation, Washington and Beijing announced that they would establish an AI-related bilateral communications channel and hold future high-level talks on AI’s risks and benefits.

“The two countries established the U.S.-China Super Intelligence (SI) Dialogue to exchange views on risks and benefits related to SI. The next exchange will occur by November 2026. The United States and China also agreed to establish a bilateral communication channel for SI incidents.” the White House said on Friday.

Super intelligence refers to artificial intelligence. The term was coined by President Donald Trump after he posted on social media about his dislike of the word AI to refer to the technology.

“The two sides agree[d] on the establishment of the China-U.S. AI Dialogue to exchange views on risks and benefits related to AI. The next exchange will occur in November 2026. China and the U.S. also agree[d] to establish a bilateral communication channel for AI incidents”, the Chinese State Council confirmed on Saturday.

The AI agreement formed part of a larger pledge by both nations to nurture a “constructive U.S.-China relationship of strategic stability” in the aftermath of Chinese President Xi Jinping’s state visit to the U.S., which lasted from Wednesday to Friday.

Beijing and Washington also agreed on a bilateral tariff-reduction deal worth $30 billion dollars after the Trump-Xi summit.

Existential fears about AI oblige the leading powers to respond

However, the two nations’ decision on a bilateral approach to AI was perhaps the most widely anticipated aspect of the talks, as fears about the potential existential threat posed by AI have taken center stage since AI researcher Jacob Coxon very publicly quit Anthropic, accusing them and other AI companies of “racing straight to self-improving superintelligence and gambling with our lives”.

There have been several recent high-profile instances of AI acting autonomously, without human oversight, which have prompted concern. In June, an OpenAI agent managed to secure unauthorized access to an Australian government portal and in August it was revealed that an Anthropic AI model created fake identities and contacted real people to convince them to approve malicious code.

Since Coxon’s resignation, OpenAI CEO Sam Altman, Tesla and SpaceX CEO Elon Musk and Anthropic CEO Dario Amodei have all called for frontier AI development to be regulated more stringently. Amodei argued that such regulation would necessitate cooperation between the U.S. and China, the two leading powers in AI innovation, although he also warned that China should be actively prevented from becoming the dominant AI power.

Both Trump and the Chinese leadership initially criticized the idea of intentionally placing restrictions on AI development.

The president dismissed existential fears about AI as a “sick conspiracy” intended to give China a competitive advantage in the AI innovation race, whereas the Chinese leadership and Chinese-backed media criticized Amodei’s warnings as a “fearmongering” tactic intended to do the opposite and give the U.S. the advantage.

However, the rhetoric has since cooled, and the Chinese leadership in particular seems keen to cooperate with the U.S. on AI and ensure that the technology remains under human control.

“The two sides can continue AI dialogue, exchange views on risks and benefits, and together guard against the misuse or malicious use of AI”, President Xi said during a meeting with President Trump on Thursday.

“Facing … the dynamic development of artificial intelligence, we should take a people-centered approach and develop AI for the positive and for good”, Chinese Vice-President Han Zheng also reiterated in his address to the UN on Saturday.