Full Breakdown
UN Security Council to Hear AI Risk Briefings Amid Global Calls for Slowdown
By Drooid · · How we work
Core Event
The 15-member United Nations Security Council will hold a public meeting on Wednesday, September 23 to discuss AI risks to international peace and security. Reuters reports Chinese firms DeepSeek and Moonshot have been invited, OpenAI CEO Sam Altman will attend in person, and Anthropic representatives are expected to brief remotely. France holds the Council presidency, with Foreign Minister Jean-Noël Barrot chairing.
Background & Context
The Council first debated AI risks in 2023, when China warned the technology could become a “runaway horse” and the United States cautioned against its use for repression. Since then, industry leaders—including OpenAI, Anthropic and Hugging Face—have called for coordinated safeguards and a slowdown of powerful models. Chinese state media have dismissed slowdown proposals as a “Cold War playbook” aimed at preserving U.S. dominance.
Key Figures & Groups
- Sam Altman – OpenAI CEO, will address the Council.
- Dario Amodei – Anthropic CEO, expected to brief remotely.
- Liang Wenfeng – DeepSeek founder, not slated to attend.
- Clement Delangue – Hugging Face CEO, listed among briefers.
- Yoshua Bengio – Co-chair, UN AI panel, expected participant.
- Scott Bessent – U.S. Treasury Secretary, announced a bilateral AI safety dialogue.
Timeline
- 2023 – First dedicated Security Council meeting on AI.
- Sept 22 – Reuters reports DeepSeek will brief the Council.
- Sept 23 – Scheduled Council meeting on AI and security.
- Early Sept – Bessent confirms U.S.–China AI safety dialogue and an “incident line.”
- ~2 months later – Planned follow-up meeting in Shenzhen.
Why It Matters
The briefing places AI safety alongside traditional security topics such as armed conflict and terrorism, signaling that autonomous weapons, information warfare and AI-linked cyber operations are now viewed as potential threats to stability. Participation by leading AI firms shows industry acknowledgment that unchecked development could outpace existing regulations, while Chinese involvement underscores the geopolitical dimension.
Official Statements & Responses
- UN Secretary-General Antonio Guterres warned that rapidly advancing AI poses risks that cannot be ignored.
- OpenAI spokesperson confirmed Altman’s focus on international coordination and safety standards.
- Scott Bessent said the United States and China will hold further talks and maintain an “incident line.”
- Jean-Noël Barrot will chair the session as France’s Security Council president.
Conflicting Reports & Gaps
Sources differ on the Chinese delegation’s composition: Reuters notes DeepSeek’s founder will not attend and the representative remains undisclosed, while TheDeepDive says Anthropic’s senior representation is “expected” but unconfirmed. No statements from DeepSeek or Moonshot have been obtained.
What’s Next
After the September 23 briefing, U.S. and Chinese officials plan to reconvene in Shenzhen in roughly two months to advance the AI safety dialogue. The Council’s outcome could shape future UN initiatives on AI governance and influence bilateral incident-reporting mechanisms.