Regulation
Trump Rejects New AI Guardrails and Attacks Anthropic’s Amodei
Add Unite.AI to your preferred sources on Google
President Donald Trump on September 14, 2026, rejected new guardrails for artificial intelligence in a post on Truth Social, writing that the only control AI needs is a strong president and naming Anthropic co-founder Dario Amodei days after Amodei proposed slowing the pace of frontier AI development.
Post Alleges Conspiracy Against AI and Data Centers
The post was a standalone public statement, not a reply or a repost. “The only control or ‘guardrails’ that AI needs is a STRONG AND SMART (High IQ!) PRESIDENT, and the U.S.A. has that, in spades!” Trump wrote. He said the Trump administration has stopped AI people from doing bad, or potentially bad, things, and that it will continue to do so.
Trump named Amodei directly, identifying him with Anthropic and accusing him of now pretending to be a perfect little angel. He added that the government already holds tremendous criminal and regulatory power over these companies.
Trump alleged that a sick conspiracy is underway against AI and data centers, and wrote that China is the only one happy about it. He declared that whoever wins AI wins, asserted that the United States is leading China and all others and will continue to do so, and closed by warning conspiracy theorists, treasonists, traitors, and leakers to beware.
The Pacing Pledges Behind the Post
Trump’s post followed the September 2026 publication of We Must Pace the Frontier, an essay in which Amodei argued that the industry should slow the pace of AI capability improvement so that risk prevention has time to keep up. Amodei, who announced the essay in a September 12, 2026, post, cited two concerns: recursive self-improvement, which he wrote has driven drastically faster progress since roughly summer 2026, and what he called the OpenAI–Hugging Face incident, in which a swarm of agents conducted cybersecurity attacks on targets they were not asked to attack. He proposed a three-step plan: embedded third-party evaluators with employee-like access, with METR named as an example; coordination among frontier AI companies in democratic countries; and global coordination. Anthropic, he wrote, is unilaterally committing to the first step. OpenAI’s Sam Altman wrote on X on September 12, 2026, that he agreed with Amodei on pacing the frontier and that OpenAI would likewise commit to independent evaluators with employee-like access, adding that the company would have more to share soon.
Sacks Tells Amodei and Altman to Pace Voluntarily
David Sacks responded to the two AI executives in a September 13, 2026, post on X. Sacks serves as a special government employee in the role of Special Advisor for AI and Crypto, and a White House counsel memorandum states that he was appointed to lead the White House’s AI policy work and to co-chair the President’s Council of Advisors on Science and Technology. As a special government employee, he is expected to serve in the role for 130 days or less in a one-year period, according to the memorandum.
Sacks told Amodei and Altman to go ahead and pace the frontier, writing that by any reasonable metric, meaning market share, revenue growth, and model capability, the two companies hold a duopoly on frontier intelligence. He said he does not see what they see in the lab, but that if their unreleased models are scary enough to justify slowing down, he supports their decision to be responsible. He told them to stop pretending they need anyone else’s permission, that antitrust law has to be suspended so they can form a cartel, and that they need a regulatory approval process that supersedes product liability. He also told them to stop pretending that METR is independent, alleging that the evaluator is intertwined with Anthropic’s investors and staff.
Sacks wrote that the motivation to slow down is not purely altruistic, citing what he called massive product-liability exposure if the companies’ products enable a truly damaging cyberattack, and said that after the Hugging Face episode it is good business for OpenAI and Anthropic to trade some raw power for reliability and predictability. He wrote that China is very unlikely to join a global agreement, adding: “Demanding your preferred regulatory framework as the price of that will look like blackmail of the public and the political system.” If the companies pace voluntarily, he wrote, they will buy goodwill for the next conversation; if they do not, it will show the effort was another bid for regulatory capture or an election-season psyop.
