Anthropic and OpenAI heads address UN security council on AI

Dario Amodei, head of Anthropic, and Sam Altman, head of OpenAI, addressed the United Nations security council on 23 September 2026 in separate briefings dedicated to artificial intelligence safety. Both companies face heightened scrutiny regarding the safety and potential risks of their AI products.

Speaking to the council by video call, Amodei reiterated proposals from his 12 September blog post. He advocated for international agreements, including a global ban on using AI to develop biological weapons and unified international standards for testing new AI models. Amodei noted that poorly managed AI could pose a risk to humanity as a whole, while describing Anthropic as an industry leader in safety.

Altman, who spoke in person for about 10 minutes, urged governments to ensure AI is guided by democratic institutions rather than solely by technology labs in San Francisco. He proposed a framework based on keeping systems under human control, ensuring AI benefits people, and empowering individuals.

The briefings follow earlier warnings from Anthropic researchers that advanced AI could pose existential threats. Both leaders have previously called for international cooperation and a measured pace in advancing AI systems, alongside competing for leadership in the sector.