OpenAI and Anthropic CEOs Urge Shared AI Safety Tests at U.N.

At a U.N. Security Council briefing, OpenAI and Anthropic CEOs urged shared AI model tests and serious-incident alerts. The Council adopted no AI rule, and the U.S. rejected a global regulator. OpenAI CEO Sam Altman and Anthropic CEO Dario Amodei urged governments at a U.N. Security Council briefing on September 23, 2026, to agree on how powerful AI systems should be tested and serious incidents reported. The two companies build the models at issue, and shared checks would give governments comparable information about their risks. The Council adopted no AI rule at the meeting.

First reported by WinBuzzer. This page is a short excerpt; the full text is theirs.