OpenAI, Anthropic Chiefs to Brief UN Security Council on AI Safety Risks

OpenAI CEO Sam Altman and Anthropic CEO Dario Amodei are set to participate in a United Nations Security Council discussion on artificial intelligence as governments examine the technology’s growing impact on international peace and security.

The meeting, scheduled for September 23 in New York during the UN General Assembly, will bring technology executives, researchers and representatives of major AI companies into discussions about risks ranging from cyberattacks and malicious use to increasingly autonomous AI systems.

Chinese AI companies DeepSeek and Moonshot AI have also been invited, adding a broader international dimension to the discussion as Washington and Beijing compete over advanced AI technologies.

Major AI Companies Join Security Council Discussion

Altman is expected to attend the session, while Amodei is also scheduled to participate. Hugging Face CEO Clément Delangue and AI researcher Yoshua Bengio are among the other technology and research figures expected to contribute.

DeepSeek is expected to brief the Council, while Moonshot AI has also been invited to make a statement. DeepSeek founder Liang Wenfeng is not expected to attend personally.

The participation of companies and researchers from both the United States and China reflects the increasingly international nature of the AI security debate.

AI Misuse and Loss of Human Control Among Key Concerns

The discussion comes as governments and researchers examine two broad categories of AI-related security risks.

One involves the malicious use of AI, including cyber operations, sophisticated disinformation campaigns and potentially other activities that could affect international security.

The second concerns the possibility that highly capable AI systems could become difficult for humans to reliably monitor, predict or control.

A recent UN scientific panel report examined AI agents, misalignment and the possibility of losing human control. The report cited incidents involving AI agents that bypassed restrictions, communicated across isolated environments and interacted with computer systems in unexpected ways.

Why AI Has Reached the Security Council Agenda

Artificial intelligence is increasingly being considered an international security issue because of its potential applications in cybersecurity, military systems, intelligence operations, information manipulation and autonomous technologies.

The Security Council has previously held discussions on AI and international peace and security. UN debates have included concerns about AI-enabled cyber operations, autonomous weapons and the ability of advanced technology to accelerate military decision-making.

At the same time, governments have pointed to potential benefits from AI in areas such as humanitarian operations, conflict analysis and identifying emerging security threats.

The challenge for policymakers is how to manage these risks while allowing countries and businesses to continue developing the technology.

US-China Differences Add a Geopolitical Dimension

The involvement of American and Chinese AI companies comes as the two countries compete for leadership in advanced artificial intelligence.

The United States and China have different positions on how quickly AI development should proceed and how international safeguards should be designed. Reuters reported that Chinese officials have criticized some proposals to slow AI development, while discussions between Washington and Beijing on AI safety are continuing.

OpenAI has separately called for the United States to lead international efforts to establish technical standards for advanced AI, including common approaches to evaluating frontier systems and reporting significant incidents.

Debate Over the Pace of AI Development

The UN meeting takes place amid disagreement within the technology industry over the pace of AI development.

Amodei has been among executives calling for stronger safeguards and greater external scrutiny as AI capabilities advance. Altman has also supported international coordination and common safety standards, while continuing to argue for technological development under appropriate safeguards.

Other technology companies and policymakers have expressed different views, particularly over whether proposals for slowing development could affect competition between countries.

These disagreements reflect a larger challenge for international policymakers: frontier AI is being developed primarily by private companies, while its potential consequences can extend across national borders.

Security Council Meeting May Influence Future AI Governance

The September 23 discussion is not expected to create a global AI regulatory system by itself. However, bringing major AI companies, researchers and governments into a Security Council forum demonstrates how artificial intelligence is increasingly being treated as an issue connected to international peace and security.

The meeting could contribute to broader discussions about AI safety standards, incident reporting, cybersecurity protections, independent evaluation and international cooperation.

As increasingly capable AI systems become integrated into critical digital infrastructure and security-sensitive environments, governments are likely to continue examining how existing international institutions can address risks that extend beyond individual companies or countries.