# U.S. and China quietly build AI safety channel

> Trump publicly rejects AI guardrails while his Treasury secretary proposes formal talks with China on managing AI risks.
- Title: U.S. and China quietly build AI safety channel · Fortune
- Summary: Trump publicly rejects AI guardrails while his Treasury secretary proposes formal talks with China on managing AI risks. The U.S. and China have begun…
- Keywords: artificial intelligence, u.s.-china relations, ai safety, trump administration, openai, anthropic, xi jinping, scott bessent, China, quietly, build, safety
- Source: Fortune — https://fortune.com/2026/09/24/us-china-ai-labs-converge-ai-guardrail-hotline
- Author: Eva Roytburg
- Read time: 2 min
- Topics: artificial intelligence, u.s.-china relations, ai safety, trump administration, openai, anthropic, xi jinping, scott bessent
## AI CEOs warn the Security Council
The CEOs of two top American AI companies told the UN Security Council on Wednesday that artificial intelligence has become a threat to global security and asked governments to write shared rules for it.
## Trump says leave AI exactly where it is
On Thursday morning, ahead of his summit with Chinese President Xi Jinping, Trump posted that super intelligence would be discussed but he wanted to leave it exactly where it is, adding that China's position was the same and that the Justice Department was the guardrail.

> Our guardrail is the DOJ!
## Xi strikes a cooperative note
Standing next to Trump at the White House, Xi said the two nations had both the capability and responsibility to develop and manage AI for good and ensure that development of AI is always under human control.

> We have both the capability and responsibility to develop and manage AI for good and ensure that development of AI is always under human control.
## Behind the scenes, talks begin
China's commerce ministry said the countries held their first dialogue on AI after Treasury Secretary Scott Bessent met Chinese Vice Premier He Lifeng in New York. Bessent proposed a notification mechanism between the two countries for national security-related AI incidents.
## Public posture and private diplomacy diverge
Trump told the UN General Assembly the U.S. rejects any globalist scheme to control AI, yet his Treasury secretary is proposing a formal channel with China, the country Trump and American venture capitalists cast as the main villain in the AI race.
## Anthropic CEO proposes phased approach
Anthropic CEO Dario Amodei called for starting with small agreements every country could agree on, such as a ban on using AI to make biological weapons, then building systems for governments to verify each other's commitments, and then a global alert system.

> No leader, no company, and no nation can manage this alone.
## Altman says OpenAI has slowed development
OpenAI CEO Sam Altman called for fast, shared incident reporting and said his company had slowed down its own AI development before and would again, undercutting the industry's usual argument that they are locked in a race they cannot escape from.
## Tech industry splits on regulation
Meta CEO Mark Zuckerberg told NBC News the industry does not need industrywide coordination, arguing there is plenty of commercial incentive to get it right. Nvidia CEO Jensen Huang has called AI fears overblown. Trump's position is much closer to theirs than to Altman's or Amodei's.
## Key takeaway

The U.S. and China have begun private talks on AI safety through a notification mechanism for security incidents, even as Trump publicly opposes international AI rules.