Dario Amodei stated on Wednesday that the trajectory of artificial intelligence could result in 'geniuses in the data center' within two years, with models potentially slipping beyond human control. During an appearance before the UN Security Council, Amodei, the head of Anthropic, and OpenAI's Sam Altman urged world leaders to establish global safety standards to mitigate the risk of the technology becoming a threat to humanity as a whole. Altman noted that OpenAI has unilaterally slowed development in the past to maintain safety and would do so again, while Amodei pledged to slow his company's pace as much as necessary.

The proposal for international coordination met opposition from the White House Office of Science and Technology Policy, which criticized the idea as a 'globalist scheme.' Director Michael Kratsios argued that the pace of intelligence advancements does not justify pausing development or creating new global governance structures. The council also heard briefings from scholar Yoshua Bengio and representatives from DeepSeek, with some officials calling for AI control measures akin to the nuclear arms treaties of the Atomic Era.