President Donald Trump postponed a highly anticipated executive order on artificial intelligence and cybersecurity just hours before it was scheduled to be signed at the White House, halting what would have been one of the administration's most significant moves on AI oversight.
The order had been expected to create a voluntary framework for AI companies to coordinate with the federal government before releasing advanced AI models and to strengthen cybersecurity protections tied to rapidly advancing artificial intelligence systems. The move came amid growing concerns in Washington over the national security implications of increasingly powerful AI technologies and their potential use in cyber operations.
Trump said he personally decided to delay the order after reviewing its content.
"I didn't like certain aspects of it. I postponed it," Trump told reporters in the Oval Office. "I think it gets in the way of — you know, we're leading China, we're leading everybody, and I didn't want to do anything to get in the way of that lead," according to Axios.
According to Axios, discussions involving Trump, AI adviser David Sacks and industry figures took place before the decision was made to shelve the order. The outlet reported that Meta CEO Mark Zuckerberg, xAI CEO Elon Musk and Sacks all spoke with Trump between Wednesday night and Thursday morning.
One source familiar with the discussions told the outlet that opposition centered on concerns that the directive amounted to unnecessary regulation. Another one told the publication that both Trump and Sacks objected to aspects of the proposal.
The delay is a setback for officials and policymakers who had pushed for additional safeguards around advanced AI systems. The order was expected to establish a voluntary process allowing developers of cutting-edge AI models to provide the government with information before public release, part of a broader effort to address cybersecurity and national security concerns surrounding the technology. Axios previously reported that the framework would seek to bolster cybersecurity around advanced AI models while creating a channel for government review of major releases.
The debate comes as artificial intelligence becomes increasingly tied to geopolitical competition, particularly between the United States and China. The Trump administration has repeatedly framed AI leadership as a national security priority. In July 2025, the White House released its "America's AI Action Plan," which emphasized accelerating innovation, expanding AI infrastructure and maintaining U.S. technological leadership. The plan described winning the AI race as critical to economic competitiveness and national security, according to the White House.
The postponed executive order also exposed disagreements over how AI oversight should be structured. Some industry officials reportedly questioned why the Treasury Department would play a prominent role in coordinating responses to AI security vulnerabilities. Critics argued that agencies such as the Cybersecurity and Infrastructure Security Agency and the National Institute of Standards and Technology have traditionally taken the lead on evaluating and disclosing critical cybersecurity risks.
Questions also remained over how the proposed voluntary testing system would operate. Technology companies have generally supported model testing and safety guardrails, and that leading frontier AI models already undergo voluntary testing through the Commerce Department's Center for AI Standards and Innovation. However, uncertainty persisted over which AI systems would be covered and whether advance-sharing requirements could complicate cooperation with allied governments conducting their own reviews.
No new signing date has been announced. According to the Axios report, officials within the White House Office of the National Cyber Director have indicated in private conversations that additional AI security initiatives remain under development beyond the postponed executive order.
