Tech chiefs persuade Trump to block proposed AI regulator

Three of Silicon Valley's most powerful executives have successfully lobbied President Donald Trump to kill off a proposed federal watchdog for artificial intelligence.

Mark Zuckerberg, Elon Musk and Nvidia's Jensen Huang made separate contacts with the president over the last week to argue against the plans, according to a report by the Wall Street Journal. The proposed body, which would have tested advanced AI systems before they were released to the public, was quietly abandoned after their intervention.

The plan for the regulator was first suggested on 14 July by Demis Hassabis, the head of Google DeepMind. Under his proposal, tech firms would have funded an independent standards and testing body—modelled on Wall Street's self-regulatory body, FINRA—to evaluate "frontier" AI models. The idea gained backing from Anthropic chief executive Dario Amodei and OpenAI boss Sam Altman.

But rivals saw it as a trap. Zuckerberg, Musk and Huang reportedly warned Trump that such a system would create a "regulatory moat", locking in the dominance of Google, OpenAI and Anthropic while punishing competitors.

Trump appeared to buy the argument. In a series of posts on his Truth Social platform, the president dismissed fears about artificial intelligence as a "hoax" cooked up by "traitors" to the United States. He claimed that any regulatory slowdown would hand a crucial advantage to China.

"The only control or 'guardrails' that AI needs is a STRONG AND SMART (High IQ!) PRESIDENT," Trump wrote, adding that existing criminal and regulatory laws were already sufficient.

The three tech leaders have since taken their arguments public. Speaking at the All-In Summit in Los Angeles, Musk argued that instead of a government regulator, leading AI companies should simply test each other's models before they launch. "Instead of grading your own homework, you would at least have competitors grading your homework," Musk said.

At the Dreamforce conference in San Francisco, Nvidia's Huang was even more blunt. "We don't need any new laws. We don't need new regulations," he said. He argued that responsibility should lie with the developers themselves: "If you build a product or a service and you're not confident in its functionality, capability or safety, then don't release it."

The stakes are enormous for all three men, who are pouring billions of dollars into AI infrastructure. Meta is currently projecting capital expenditure of up to $145 billion for 2026, while Musk's SpaceX is rapidly scaling up its AI compute segment, aiming to bring 20 gigawatts of power and cooling online by the end of next year. Nvidia, which supplies the essential chips powering these projects, is struggling to keep up with demand, currently meeting only about 70 percent of its orders.

With the federal watchdog plan dead, focus now turns to how White House advisor David Sacks will shape future AI policy. Sacks is known to favour lighter, voluntary guidelines over hard regulations.