SKN | OpenAI and Anthropic CEOs to Brief UN Security Council on AI Risks and Global Security

Key Points

• OpenAI CEO Sam Altman and Anthropic CEO Dario Amodei are expected to brief the UN Security Council on artificial intelligence and its implications for international security.
• Chinese AI companies DeepSeek and Moonshot have also been invited, broadening the discussion to include competing approaches from the United States and China.
• The meeting comes as governments and AI companies debate how to manage increasingly capable systems, including concerns around misuse, autonomous behavior and the potential difficulty of maintaining human control.

Artificial intelligence is moving deeper into the international security agenda, with the leaders of two of the world’s most prominent AI companies preparing to address the United Nations Security Council as governments examine the risks associated with increasingly capable systems.

OpenAI CEO Sam Altman and Anthropic CEO Dario Amodei are expected to participate in the Security Council’s Wednesday meeting, alongside Hugging Face CEO Clément Delangue and AI researcher Yoshua Bengio, co-chair of the UN’s Independent International Scientific Panel on AI. The United Nations has confirmed the participation of the technology executives.

AI Moves Further Into International Security Debate

The meeting reflects a shift in how advanced AI is being treated by governments. Rather than remaining primarily a technology-policy issue, the development of increasingly capable models is being examined through the lens of international stability, cybersecurity and potential misuse.

The Security Council first formally discussed artificial intelligence in 2023. The latest meeting comes as AI systems have become substantially more capable and as researchers, technology executives and policymakers debate whether existing safeguards are sufficient.

France is convening the session during its September presidency of the Security Council. Reuters reported that the meeting will focus on the implications of AI for international security, including concerns surrounding the rapid advancement of increasingly autonomous systems.

Chinese AI Firms Expand the Scope of the Discussion

The participation of Chinese AI companies adds another dimension to the meeting. DeepSeek and Moonshot have been invited to make statements, according to Reuters. DeepSeek is expected to participate, although founder Liang Wenfeng is not expected to attend.

That participation places major US and Chinese AI developers within the same international discussion at a time when competition between the two countries over advanced AI capabilities is becoming increasingly important to technology and national-security policy.

The inclusion of companies from both sides also underscores the difficulty of establishing common standards for AI development when governments and technology companies have different priorities surrounding safety, competitiveness and technological leadership.

Amodei’s Call for Greater AI Safety Coordination

The Security Council meeting follows a series of increasingly public warnings from AI executives about the pace of technological development.

Amodei recently called for AI companies to slow the rate at which they improve model capabilities, arguing that additional time is needed to develop safeguards and independent evaluation mechanisms. His proposal included greater cooperation among frontier AI developers, independent evaluators with significant access to AI systems and stronger international coordination. OpenAI’s Altman has expressed support for independent evaluation measures.

Anthropic and Accenture have subsequently announced plans to invest at least $2 billion over five years in independent evaluation of frontier AI models, including red-teaming and safety assessments.

OpenAI Pushes for International Technical Standards

OpenAI has also called for the United States to lead international efforts to establish technical standards for advanced AI, including systems capable of recursive self-improvement.

The company has argued for common standards and incident reporting mechanisms while supporting what it describes as “pacing” AI development so that safety research can keep up with advances in model capabilities. Altman’s appearance before the Security Council comes against this broader backdrop of discussions over international coordination.

At the same time, governments and technology companies remain divided over how such safeguards should be implemented and who should ultimately oversee them.

Outlook

The UN Security Council discussion places AI development directly within the broader framework of international security and governance. The participation of OpenAI, Anthropic and Chinese AI companies could provide a rare forum for comparing views on model safety, international standards and the risks associated with increasingly autonomous systems. The outcome may not immediately produce binding rules, but the meeting signals that advanced AI is increasingly being considered an issue with implications extending beyond the technology sector.