OpenAI and Anthropic CEOs Brief UN Security Council on AI Safety

The heads of two leading artificial intelligence companies made their case for AI governance to the UN Security Council on Wednesday, presenting contrasting visions for managing the technology's risks while both calling for greater international oversight.

Sam Altman of OpenAI and Dario Amodei of Anthropic delivered separate briefings as their companies face mounting scrutiny over AI safety. The appearances came weeks after Anthropic researchers issued warnings that AI could pose existential threats to humanity within the decade, prompting industry leaders including both CEOs to advocate for slowing development and establishing international safety standards.

Altman's Democratic Vision Versus Lobbying Reality

Speaking from the security council's horseshoe table, Altman framed AI's trajectory as a choice between "a new renaissance of creativity and discovery, or more like a new industrial revolution of upheaval and disarray." He emphasized that critical AI decisions "cannot be made by labs in San Francisco alone" and must be shaped through democratic processes and accountable governments.

Altman outlined three principles: AI systems must remain under human control, benefit people from healthcare to entrepreneurship, and empower individuals. He cautioned against "doomerism" while acknowledging safety concerns, advocating instead for a "middle path."

The speech, however, stood in tension with OpenAI's political activities. The company's president Greg Brockman leads the Leading the Future PAC, which has raised over $100 million and spent millions targeting candidates who support stronger AI regulation.

Amodei Doubles Down on Existential Risk

Amodei, appearing via video call, took a more direct approach to catastrophic risk. "If managed poorly, I even believe that AI could be a risk to humanity as a whole," he told the council, while asserting that Anthropic has "the best track record in the industry for preventing these risks."

He advocated for specific international agreements, including bans on using AI to develop biological weapons and global standards for testing new AI models. His remarks echoed points from a September 12 blog post calling for increased global cooperation.

Why It Matters

The UN briefings represent a critical juncture where AI safety discourse has moved from industry forums to international security bodies. Both companies are positioning themselves as responsible stewards while racing toward potentially record-setting IPOs that could value each above $1.5 trillion. The tension between their public calls for regulation and their competitive positioning—along with OpenAI's political spending against regulatory candidates—raises questions about whether voluntary industry leadership can address risks that researchers within these same companies describe as existential. The involvement of the UN Security Council signals that AI governance is now treated as a matter of international security rather than purely technological innovation.

Competing for Safety Leadership

Both CEOs have long sought to shape public perception of AI while their companies compete for market dominance. The addresses come as AI regulation has climbed the political agenda following Anthropic's stark warnings about technology potentially wiping out humanity. These concerns, while debated within the AI community, center on scenarios ranging from bioweapon development to financial catastrophe.

Earlier this month, Amodei's call for an AI development slowdown received support from OpenAI, Google, and Elon Musk, though the practical implementation of such restraint remains unclear as companies continue aggressive development timelines.

The details were first reported by The Guardian, with Nick Robins-Early and Dan Milmo covering the Security Council briefings.