Trump calls AI risk a 'hoax' and lumps Amodei's warnings in with climate change

By Zendoric · September 15, 2026.

Donald Trump has posted a message on social media in which he calls himself a "Hoaxbuster" and places three very different things on the same plane: the risk of artificial intelligence spinning out of control, climate change, and the accusations of Russian interference in US elections. As reported by Demócrata, Trump claims that "the people who say AI is going to destroy the world and that data centers are bad for your neighborhood are the same ones who were saying not long ago that the world would go extinct from 'climate change.'" He describes those who warn of the risk as "evil revolutionaries" who, in his view, "work for people who do not have the interests of the United States in mind."

The message does not come out of nowhere. It arrives, according to the article itself, amid the public debate opened by Dario Amodei, co-founder and chief executive of Anthropic, who has called for "slowing down" the development of AI model capabilities given the risk of "losing control" of the technology and of triggering "cyberattacks," "bioterrorism" or "serious economic problems." The New York Times, which has covered the same episode, headlines outright that Trump "rejects AI regulation and singles out Anthropic's chief executive," confirming that the barb has a name attached.

It is worth separating the rhetoric from the argument. Comparing warnings about AI risks to the climate change "hoax" is a political maneuver, not a technical rebuttal: it offers not a single piece of data contradicting what Amodei is arguing, it merely reclassifies it as a hoax by association. And the nuance matters because Amodei is neither a marginal voice nor an outsider to the business: he runs one of the world's leading AI labs, whose business model depends on the industry continuing to advance at full speed. That he is calling for a slowdown runs against his own immediate commercial interest, which makes his warning more credible, not less.

The backdrop is as much industrial policy as security. A White House that dismisses the risk as a hoax fits with an agenda of accelerated data center deployment, cheap energy to train models and minimal regulatory friction; slowing down now is read, in that calculus, as ceding ground to China in the race for the AI frontier. The problem is that governing the technology well and maintaining leadership are not opposing goals: a serious cybersecurity incident or a biological misuse — precisely the risks Amodei cites — would damage the US position far more than any well-calibrated regulatory brake.

At Zendoric we hold that the near-term risk is real and is not resolved by labeling it a hoax: cybersecurity, biological misuse and the economic disruption that agentic AI can cause are concrete problems that demand evidence-based governance, not political denial or panic. But that same honesty about the short term is compatible with our underlying thesis: AI, well governed, is the most plausible path to a society with less disease, better health and an abundance of resources. What is at stake in this episode is not whether AI is good or bad, but whether the United States builds the safeguards now, while it still has a technological edge, or waits for a serious incident to force them upon it. Dismissing as a "hoax" those who warn of the second scenario does not make it any less likely.