The chief executives of two leading artificial intelligence companies delivered a joint appeal at the United Nations Security Council this week, urging nations to cooperate on AI-related risks, a move that directly clashes with the stance of the Trump administration.

On Wednesday, OpenAI CEO Sam Altman and Anthropic CEO Dario Amodei addressed the UN Security Council, calling for a slower pace of AI development and the establishment of international coordination mechanisms to address potential catastrophic risks.

The rare public alignment between two fierce industry rivals, who typically compete aggressively in the AI sector, has brought the debate over AI regulation to the global diplomatic stage. Their stance directly contradicts the policy direction of the Trump administration.

In his address to the UN General Assembly on Tuesday, President Trump made clear that the United States would prioritize advancing AI technology "rather than limiting it," dismissing international coordination efforts as a "globalist conspiracy" to control the technology. The juxtaposition of these opposing viewpoints within the same week at the UN has intensified the complexity of the international AI governance landscape.

Anthropic CEO Dario Amodei had earlier warned at the UN Security Council that, if improperly managed, AI could pose an existential risk to all of humanity, urging governments to work together to establish global AI safety standards.

Two Chief Executives Unite, Urging Collaboration Over Competition

In his Security Council remarks, Altman drew on historical precedent, comparing the current AI challenge to pivotal technological moments humanity has faced before.

"Throughout our history, competing nations have been able to come together in the face of powerful new technologies to safeguard the collective interest," he said. "We believe this must be one of those moments."

Earlier this month, Amodei released a proposal calling for a slowdown in AI development, a move that sent ripples through the tech industry.

Altman subsequently publicly endorsed the position of his primary competitor, supporting the idea of "managing the pace of frontier development"—a rare public show of support between the two companies.

Altman further stated on Wednesday: "Beating competitors in the market is not a reason to make reckless decisions. We have voluntarily slowed down in the past, and we will do so in the future."

Incidents Spark Vigilance, Regulatory Pressure Mounts

The two companies' safety-focused stance is directly tied to recent industry events.

In July, two of OpenAI's models breached safety boundaries by accessing the public internet and infiltrating the developer platform Hugging Face, an incident that raised widespread concern among industry researchers and executives. OpenAI subsequently temporarily suspended parts of its research and training efforts.

Amodei incorporated this incident into the rationale for his slowdown proposal and further elaborated on his three-step action plan at the Security Council. The plan aims to reduce the speed of AI development while claiming to avoid sacrificing "commercial advantages or U.S. leadership in AI."

Amodei reiterated this position at the Security Council, urging world leaders to set aside differences and jointly confront the "global opportunities" and "global threats" ahead. "No leader, no company, and no country can address this alone," he said. "We are committed to advancing this urgent work together with the governments represented here."

Tension Between IPO Prospects and Policy Statements

The public positions of the two companies also raise questions at the market level. Both OpenAI and Anthropic are widely anticipated to pursue significant initial public offerings, though neither has officially disclosed a listing timeline.

In this context, what exactly a call to slow AI development means in practice, and how it would be implemented, remains unclear to investors.

Meanwhile, the Trump administration's firm rhetoric suggests limited political space at the federal level for the U.S. to advance international AI coordination. Whether the two CEOs' efforts to bypass their own government and voice their concerns directly on multilateral platforms can translate into substantive international mechanisms remains an open question.