President Trump on Monday rejected appeals to impose additional safety controls on AI models, dismissing the concerns of industry leaders who have urged a slowdown in frontier AI development.
In a post on Truth Social, Trump declared that his administration already holds "very strong criminal and regulatory power!" over AI companies, while claiming there is a "conspiracy" against AI and data centers.
His comments came after Anthropic CEO Dario Amodei, OpenAI executives, and Elon Musk of SpaceX's xAI all publicly called for a more measured pace in developing cutting-edge AI models.
Trump wrote that the only "control" or "guardrail" AI needs is "a very powerful and smart (high IQ!) President, and we have exactly that in the USA!" He added that "the Trump Administration has stopped AI insiders from doing bad or potentially risky things, just like Anthropic's Dario, who is now pretending to be a perfect little angel. We will continue to do so!"
Earlier this year, Anthropic clashed with the Trump administration after the company refused to provide technology to the U.S. Department of Defense for mass surveillance of American citizens or for the development of fully autonomous weapons. In response, Trump and Secretary of Defense Pete Hegseth labeled the firm a supply chain threat and barred other government contractors from collaborating with the AI startup.
Over the weekend, Amodei published a lengthy essay outlining steps the industry should take to slow AI development, including embedding third-party observers within companies and supporting federal AI regulations.
For years, many in the tech sector have warned about AI's potential dangers. The latest wave of concern was triggered by two AI researchers who publicly cautioned about the hidden dangers of advanced large language models.
A former Anthropic researcher publicly criticized the company and OpenAI for failing to take responsibility in developing high-performance AI systems. In a post on X, the researcher, identified as Coxon, stated that both firms are racing to build superintelligence capable of self-improvement.
"No other human activity carries such a high level of risk," he wrote.
Shortly after, Evan Hubinger, Anthropic's head of alignment science, responded to Coxon's post, agreeing with the assessment. He estimated that there is more than a 10% probability that AI technology will eventually "lead to the extinction of all humanity."