Trump Rejects AI Safety Warnings at UN; 4 Industry Leaders Disagree

For the AI community, Trump's September 22 UN speech was not just another political moment—it was a direct dismissal of a safety position now held by the field's own leaders. When Dario Amodei, Sam Altman, Demis Hassabis, and Elon Musk align on slowing capability development, the claim that AI doom warnings are merely a political talking point collapses.

On September 22, 2026, President Donald Trump used his address to the United Nations General Assembly to defend the nearly seven-month U.S. military campaign in Iran and to dismiss mounting warnings about artificial intelligence as the product of the same people he derided as authors of the "Russia, Russia, Russia hoax" and "climate change" alarmism. A fact-check by PolitiFact, republished across multiple Hearst stations, found the president's framing misleading on the central point: concerns about AI risk are not confined to one political or activist group but are being voiced by the most influential leaders inside the AI industry itself.

“

According to the PolitiFact account, OpenAI CEO Sam Altman, Google DeepMind Co-founder and Chair Demis Hassabis, and xAI founder Elon Musk agreed with Amodei—a rare concurrence in a hyper-competitive industry.

The president's AI claim is significant less for what it says about politics and more for what it reveals about the emerging fault line between U.S. federal policy and the frontier research community. Trump told the General Assembly that the United States "totally rejects any attempt to construct a globalist scheme to control artificial intelligence." In doing so he cast international AI governance as an ideological project rather than a safety question. That stance has concrete implications. It suggests Washington may resist or refuse participation in emerging multilateral frameworks on AI safety, even as the technology's leading builders warn that capability growth is outpacing safety guarantees.

The fact-check anchors its rebuttal in a striking development: Anthropic CEO Dario Amodei published a September 12 essay urging a slower pace of AI capability development. According to the PolitiFact account, OpenAI CEO Sam Altman, Google DeepMind Co-founder and Chair Demis Hassabis, and xAI founder Elon Musk agreed with Amodei—a rare concurrence in a hyper-competitive industry. The agreement of four top executives, who otherwise compete on frontier model performance and commercial deployment, is a data point the Trump speech glosses over. It undermines the president's narrative that AI doom warnings are a partisan or ideological construct. Instead, they have become a mainstream, cross-company concern among the very people building the systems.

The political context of the speech matters too. Trump addressed the General Assembly amid a war with Iran that has stretched nearly seven months and affected fuel prices. Polls show Americans tiring of the conflict. In the same address, the president said the U.S. was "settling years of unfinished business. And we're doing it well." The conflation of military operations and AI policy is not incidental; the speech frames both as resisting international overreach. Cuban and Iranian delegations walked out at points, underscoring the hostility of the setting. Iranian President Masoud Pezeshkian was due to address the General Assembly the next day, an unusual move for a leader of a country at war with the United States.

The fact-check also notes that U.S. lawmakers from both parties have introduced legislation to regulate AI. That detail complicates the president's "globalist scheme" framing. If anything, the regulatory impulse is coming from inside the American political system as well as from abroad. The presence of bipartisan bills suggests that even domestic legislators see AI as something that cannot be left to uncoordinated market forces. Trump's UN remarks may therefore put the executive branch at odds not only with global institutions but also with Congress and the AI industry's own leaders.