OpenAI, Anthropic and Hugging Face Leaders Speak to UN on AI Threats

Executives from leading AI firms will brief the United Nations Security Council this Wednesday amid mounting international concern that rapidly advancing artificial intelligence could soon develop capabilities that exceed human control and pose security risks.

Sam Altman, CEO of OpenAI; Dario Amodei, CEO of Anthropic; and Clément Delangue, co-founder of Hugging Face, are scheduled to address the 15-member council during the UN General Assembly. Their appearances come as policymakers and diplomats weigh how best to monitor, measure and govern increasingly powerful AI systems.

Altman intends to call on world leaders to agree on standardized benchmarks for gauging AI capabilities and for evaluating safety measures companies put in place during development, an OpenAI spokesperson told reporters. Such benchmarks would aim to create clearer, comparable metrics so states and firms can assess risks more transparently.

The Security Council session will include statements from member states, including the United States and China—two leading powers and competitors in AI development. French Foreign Minister Jean-Noël Barrot will chair the briefing. A European diplomat, speaking anonymously, said the meeting’s primary goal is to highlight concrete risks stemming from AI systems getting out of control in ways that could threaten national and international security.

“Questions at the core of the council’s mandate include whether autonomous systems interacting across borders might inadvertently trigger escalation or conflict,” the diplomat said. The concern is not merely hypothetical: experts worry about scenarios in which algorithms coordinating with one another, or making rapid decisions, could produce unintended outcomes with geopolitical consequences.

As world leaders gather in New York, bilateral talks are also taking place. Ahead of a planned meeting between U.S. President Donald Trump and Chinese leader Xi Jinping, officials from Washington and Beijing have discussed creating a notification system. That framework would enable the two countries to alert one another about AI-related incidents that rise to the level of national security threats—part of broader efforts to reduce the risks of miscalculation or accidental escalation tied to AI.

President Trump, speaking to the General Assembly, emphasized a different stance: he argued against global attempts to tightly control AI, calling the technology “amazing” and insisting the United States must preserve its leadership. “I’m not going to stifle growth of something that will be bigger than the Industrial Revolution,” he said, adding that the U.S. would nonetheless “watch it closely through the Department of Justice.” His comments reflect a tension in international approaches: some leaders push for strong, coordinated governance while others prioritize innovation and competitiveness.

That view contrasts with UN Secretary-General António Guterres, who has repeatedly warned about the rapid pace of AI development and its attendant dangers. Guterres has advocated for global governance mechanisms and in 2023 formed a 39-member advisory body composed of industry leaders, officials and academics from across the world. The following year, the UN General Assembly established an Independent International Scientific Panel on AI to provide expert assessment; Yoshua Bengio, a prominent AI researcher and panel co-chair, will also brief the Security Council.

The Security Council first convened on AI risks in 2023. At that meeting, China cautioned that AI should not become a “runaway horse,” while the United States warned against the technology’s potential use in repression or censorship. In 2024, then-Secretary of State Antony Blinken chaired another council discussion on the topic, underscoring growing recognition among global security institutions that AI is now a strategic concern—not only for economic or ethical reasons but for national and collective security.

Key questions before the council now include how to define thresholds for dangerous AI behavior, which entities should be responsible for testing and oversight, and how to establish international protocols for incident reporting and response. Proposals range from developing standardized tests and red-team evaluations to establishing cross-border notification systems for AI incidents with security implications.

Industry participation in the Security Council briefing signals a willingness by major AI companies to engage with governments and multilateral institutions, even as differences remain over the best path forward. The companies’ testimony will likely emphasize technical safety measures, internal governance practices and the need for shared standards. Whether that will produce binding agreements or practical mechanisms for crisis prevention remains uncertain.

As the UN deliberations continue, diplomats and experts say the stakes are high: without clearer rules, assessments and cooperation, the speed and complexity of AI development could outpace the ability of national and international institutions to manage risks. The Security Council briefing aims to move the conversation from abstract warnings to actionable steps that might prevent AI-related incidents from creating diplomatic crises or threatening international stability.

Read more : - Nasdaq Reaches New High as Tech and AI Optimism Outweigh Geopolitical Fears