In a rare joint appearance before the United Nations Security Council, the chief executives of two leading artificial intelligence companies made a unified plea for international cooperation on AI safety. OpenAI CEO Sam Altman and Anthropic PBC CEO Dario Amodei addressed the council during a session focused on the existential threats that advanced AI systems could pose to humanity.

Both executives underscored the critical need for globally aligned standards to manage the potential for AI systems to develop self-improvement capabilities that surpass human control. They argued that governments must collaborate closely with industry leaders to make pivotal decisions shaping the technology's future trajectory.

Altman highlighted the historical precedent for such collaboration, stating that even rival nations have previously united in the face of powerful new technologies for the sake of collective well-being. He expressed confidence that the current moment calls for a similar spirit of cooperation, noting the urgent need for a shared path forward despite international differences.

The Security Council meeting convened on Wednesday during the UN General Assembly in New York, where leaders and senior diplomats debated the dangers associated with artificial intelligence. UN Secretary-General Antonio Guterres opened the session by urging world leaders to align on regulations for frontier AI models, positioning the United Nations as the platform to support such efforts.

In contrast, former US President Donald Trump rejected calls for international agreements that would impose safety guardrails on AI advancement. During his address, he declined to support formal frameworks aimed at restricting development.

Speaking via video link, Amodei acknowledged the immense potential of AI to revolutionize science and medicine but warned of catastrophic risks if left unmanaged. He pointed to the possibility of terrorists using AI to engineer lethal biological weapons and the danger of models becoming too powerful for their developers to control.

Amodei called on global powers to set aside their disagreements, emphasizing that no single leader, company, or nation can tackle this challenge alone. He framed AI as both a global opportunity and a global threat that demands urgent collective action.

Altman struck a balanced tone in his remarks, advocating for a pragmatic middle ground between those who predict AI doom and those who downplay the risks entirely. He stressed the importance of rapid-response mechanisms and international consensus on evidence, safeguards, and oversight, while also acknowledging that his own knowledge is far from complete.

He expressed openness to better suggestions from other nations and reiterated his commitment to working collaboratively to identify the safest and most effective direction for the future of AI.