Trump calls AI warnings a hoax as Beijing calls Amodei's essay fearmongering

The first was about speed. "We must slow the pace at which we improve the capabilities of AI models," he wrote, and separately that "we must pace the frontier". He has warned that within six to twelve months AI agents could be capable of taking over much of the internet, and committed Anthropic to giving third-party evaluators employee-like access to verify its safety practices.

The second was about China. Amodei wrote that a "Chinese lead in AI would pose grave danger for the United States and the world", and called for continued restrictions on sales of advanced AI chips and chipmaking equipment to China.

Washington's answer

President Donald Trump rejected the first part.

"AI taking over the World, destroying Humanity, and all other things bad, is a HOAX," he posted on Monday. He added: "Don't kill the Golden Goose!" and said there was a "SICK conspiracy going on against AI".

On the question of oversight he was equally direct, writing that the only guardrails AI needs are "a STRONG AND SMART (High IQ!) PRESIDENT".

Speaking earlier at his golf resort in Ireland, he described the warnings as exaggerated and attributed them to unspecified "negative forces", while framing the competitive stakes in terms he has used before: whoever wins AI, wins.

Beijing's answer

China rejected the second part.

Guo Jiakun, a spokesperson for China's Ministry of Foreign Affairs, said: "Fearmongering, confrontation and vicious competition will only disrupt the process of global AI governance."

The timing sharpened it. Amodei's essay appeared days after the FBI, the National Security Agency and the Cybersecurity and Infrastructure Security Agency issued a joint advisory saying Chinese AI developers had made "aggressive, malicious" efforts to extract, or distil, capabilities from advanced American models including Anthropic's Claude and OpenAI's GPT.

China's Commerce Ministry dismissed those claims as groundless, said distillation is commonly used by many AI companies, and accused the United States of seeking a monopoly of the AI industry.

Where the industry stands

Amodei was not alone among AI executives, though the others framed it more narrowly.

Sam Altman of OpenAI supported the principle while drawing a boundary around it. "When we talk about 'pacing', we do not mean 'stopping'," he said. He called for consistent safety requirements for frontier AI and proposed independent auditors, and argued that "no amount of American competitive pressure should justify recklessness".

That is a narrower proposition than Amodei's: process and verification rather than deceleration.

Three positions, not two

The result is a three-way split that does not divide along the lines the debate is usually described in.

The companies building the most capable systems are asking for constraints on themselves, with varying enthusiasm. The American government says the risk those companies describe is fabricated and that regulation would damage a national asset. The Chinese government says the risk talk is a cover for export controls aimed at slowing it down.

Amodei's essay tried to hold both halves of his argument together — slow down, and keep chips from China. Each government took the half it objected to and rejected that, leaving the other untouched. Trump did not dispute the China policy. Beijing did not dispute that the technology carries risk.

What follows

Anthropic's commitment to third-party evaluators is the one concrete measure to come out of the week, and it binds only Anthropic. Altman said OpenAI would do the same.

Vice President JD Vance, House Speaker Mike Johnson and House Democratic leader Hakeem Jeffries have all engaged with the question in recent days. No legislation has been proposed, and the administration's position as stated by the president is that none is needed.