Breaking

OpenAI and DeepSeek Set for UN Security Council AI Briefing as UNGA Puts Safety on Agenda

September 22, 2026 at 07:42 EDT

Policy & Regulation
Security
Foundation Models

OpenAI CEO Sam Altman and Chinese AI startup DeepSeek are set to brief the 15-member UN Security Council on September 23, 2026, turning AI safety into a high-level security issue during the UN General Assembly week in New York. The meeting puts frontier-model risk, US-China competition and proposed global standards into the same diplomatic frame as wars in Iran and Ukraine.

Sept. 23, 2026 · UN Security Council

AI safety reaches the security table

OpenAI, DeepSeek and diplomats are converging on a Security Council debate over frontier-model risk, shared standards and US-China incident channels.

Security Council members in the AI session

Experts on the UN scientific AI panel

26%

Claude-led share of Anthropic model R&D

Institutional scale

Security Council

UN AI panel

A diplomatic council is now drawing on a larger scientific review process.

What negotiators are trying to connect

Model incident

→

Notification channel

→

Shared standards

→

Human safeguards

Why supporters want it

Common testing, incident reporting and independent assessment could make frontier AI risks easier to compare and contain.

Why critics worry

Geopolitical rivalry, weak enforcement and incumbent-friendly rules could limit what a briefing can actually change.

Continue reading

The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.