The AI CEOs briefed the UN Security Council on losing control. The White House answered: full speed.

Today, for the first time, the UN Security Council sat down with the CEOs building frontier AI and asked them — on the record — whether their own machines might slip out of human control. Sam Altman showed up in person. Dario Amodei joined by video. Hugging Face's Clément Delangue came too, plus Yoshua Bengio, one of the actual godfathers of AI. The message was the one they've been repeating for weeks: systems are getting more capable faster than the safeguards, and the industry should deliberately pace itself while alignment research catches up. Altman plans to push world leaders to adopt real benchmarks — ways to measure what these systems can do, and whether the guardrails around them actually hold. A European diplomat gave the plain-English version of the nightmare: algorithms interacting with each other, triggering a war nobody ordered. And one day earlier, in the same United Nations building, the President of the United States told the General Assembly he "rejects any attempt to construct a globalist scheme to control" AI. Encourage it, don't rein it in. He has called the takeover warnings a hoax. So here's the edge case nobody gamed out: the people building the frontier and the government hosting the frontier walked into the same building this week — with opposite plans. Sources: Reuters; TBS News; WebPro News; USA Today.