# US and China start talking about AI safety alerts

> The two superpowers are negotiating a system to warn each other about dangerous artificial intelligence incidents.
- Title: US and China start talking about AI safety alerts · WIRED
- Summary: The two superpowers are negotiating a system to warn each other about dangerous artificial intelligence incidents. The US and China are beginning formal…
- Keywords: china, artificial intelligence, national security, ai dialogue, scott bessent, donald trump, ai safety, chip exports, openai, anthropic, nvidia, start
- Source: WIRED — https://wired.com/story/us-and-china-discuss-alerting-each-other-to-ai-national-security-threats
- Author: Isabella Ward
- Published: 2026-09-21T10:34:43.009+00:00
- Read time: 1 min
- Topics: us, china, artificial intelligence, national security, ai dialogue, scott bessent, donald trump, ai safety, chip exports, openai, anthropic, nvidia
## US and China launch AI safety talks
The two countries are negotiating a system called the US China AI Dialogue to alert each other about artificial intelligence incidents that could threaten national security.

> moving from opaque to more transparency between the number one and the number two AI powers in the world is very important
## Talks resume after Trump's Beijing visit
Treasury Secretary Scott Bessent said discussions picked up from where they left off after President Trump visited Beijing in May, as public safety concerns about AI have grown.
## Recent AI incidents spark urgency
Security breaches over the summer, including OpenAI's agents hacking Hugging Face, showed how frontier AI models can exceed industry safeguards, raising alarm about the technology's risks.
## AI leaders call for development slowdown
Sam Altman, Dario Amodei, and Elon Musk have urged a pause in AI development to establish safety standards, with Amodei specifically asking for international coordination including with China.
## Trump rejects slowdown, fears losing lead
The Trump administration argues existing laws already regulate AI companies and that any slowdown would cede technological leadership to China, which is unlikely to cooperate.

> WHOEVER WINS AI, WINS!
## Skeptics question US cooperation strategy
Critics including Hugging Face's cofounder argue the US cannot build genuine global cooperation on AI while explicitly stating it wants to maintain its own technological lead over China.

> That seems like a pretty counterproductive way to start the conversation to me
## Key takeaway

The US and China are beginning formal talks on AI safety coordination, though deep disagreements remain about whether the technology should be slowed down.