Trump Postpones AI Security Executive Order After Industry Pushback
Similar Articles
Trump Administration Signs AI Safety Agreements with Tech Firms, Executive Order May Follow
AI Safety and Trade Talks Shape Agenda for Trump-Xi Summit
White House Nears Agreement on AI Safety Oversight Framework
White House Prepares Executive Order on AI Safety and Cybersecurity
Trump Allies Urge Pre-Release Testing for Powerful AI Models
President Donald Trump postponed the signing of a new executive order on artificial intelligence (AI) and cybersecurity hours before a scheduled ceremony. He cited concerns that the order could hinder U.S. leadership in AI, particularly against China. The White House has indicated it is working on additional AI security initiatives beyond the postponed order.
Facts First
- President Trump postponed signing an AI and cybersecurity executive order hours before a scheduled Oval Office ceremony.
- Trump cited concerns the order could hinder U.S. AI leadership, stating he 'didn't want to do anything to get in the way of that lead.'
- Industry leaders including Elon Musk and Mark Zuckerberg reportedly helped derail the order, according to Semafor.
- The postponed order included a provision for government testing of frontier AI models for up to 90 days before release.
- The White House's Office of the National Cyber Director is working on additional AI security initiatives beyond the expected executive order.
What Happened
President Donald Trump called off a signing ceremony scheduled for Thursday, May 21, 2026, for a new executive order on artificial intelligence (AI) and cybersecurity. The postponement was announced hours before the event, which was to take place in the Oval Office with major tech, AI, and cyber CEOs invited. Trump told reporters he postponed it because he 'didn't like certain aspects of it' and felt it could get in the way of U.S. leadership in AI. According to reports from Axios and Semafor, industry representatives including Meta CEO Mark Zuckerberg, xAI CEO Elon Musk, and former AI adviser David Sacks spoke with Trump ahead of the signing and helped derail it.
Why this Matters to You
The push for government action to vet powerful AI systems follows concerns within industries like banking about AI's ability to identify cybersecurity vulnerabilities in software. A postponed executive order granting the government power to test frontier AI models before release could affect the pace and security of new AI tools you might use. The Cybersecurity and Infrastructure Security Agency (CISA) and the National Institute of Standards and Technology (NIST) typically lead such security reviews, but leading frontier models currently conduct only voluntary testing. The White House's ongoing work on additional AI security initiatives suggests new regulations may still be forthcoming, which could shape how AI is developed and deployed in your workplace and daily life.
What's Next
The event has been postponed to a later date, according to a note viewed by Axios. The White House's Office of the National Cyber Director has indicated in private conversations that it is working on additional AI security initiatives beyond the expected executive order. This suggests the administration may still pursue regulatory measures for AI security, though their form and timing are now uncertain. The continued voluntary testing regime through the Commerce Department's Center for AI Standards and Innovation is likely to remain the primary oversight mechanism for frontier AI models in the interim.