Trump Signs AI Executive Order Focused on Cybersecurity Oversight
President Donald Trump signed a cybersecurity-focused artificial intelligence executive order Tuesday that directs federal agencies to expand oversight of advanced AI systems while maintaining a largely hands-off approach to model releases.
The order requires developers of advanced AI to grant the U.S. government and critical infrastructure operators 30 days of pre-release access to new models. An earlier draft had proposed 90 days of early access before the signing was postponed last month due to industry concerns about overregulation.
The final version explicitly prohibits licensing or preclearance requirements, meaning the government gains visibility into new systems without formal approval authority over their release.
What Agencies Must Do Within 30 Days
- The Defense Department and other national security agencies must secure their networks
- Federal civilian agencies must secure their networks and facilitate access to advanced AI models
- The Office of Management and Budget must identify existing federal grant funding to support AI vulnerability-detection efforts
Critical infrastructure operators-including hospitals, banks, utilities, and state and local governments-must also meet the 30-day network security deadline.
Establishing a Framework for "Frontier" Models
The order creates a classified benchmarking process to determine which AI systems qualify as "covered frontier models" that warrant federal attention. The White House Office of Science and Technology Policy, the National Institute of Standards and Technology, and the Cybersecurity and Infrastructure Security Agency have 60 days to establish this evaluation process.
The National Security Agency will formally determine which systems meet the threshold. The Commerce Department will assist in developing a classified benchmarking framework that informs a voluntary system for AI developers.
Building Government AI Capacity
The Treasury Department, working with the Office of the National Cyber Director, the NSA, and the Department of Homeland Security, must establish a voluntary coordination clearinghouse between government, AI companies, and critical infrastructure operators.
The Office of Personnel Management has 60 days to increase cyber hiring through the U.S. Tech Force. The force, launched in December, has onboarded 10 employees so far and has been recruiting cyber talent in recent weeks.
Why the Shift in Approach
The administration's strategy reflects concerns about recent AI advances in cybersecurity. Anthropic's Mythos model and OpenAI's GPT-5.5-Cyber have demonstrated sophisticated capabilities for uncovering network vulnerabilities, prompting officials to consider how these systems could reshape both defensive and offensive cyber operations.
The order attempts to balance federal visibility over advanced AI development with industry concerns about regulatory burden. The reduced pre-release window and absence of formal approval requirements signal the administration's preference for monitoring over control.
For government employees, this order establishes new responsibilities across cybersecurity, infrastructure protection, and AI governance. Agencies now face compressed timelines to secure systems and coordinate with the private sector on AI access and development standards. Learn more about AI for Government and AI for Cybersecurity Analysts to understand how these systems will affect your work.
Your membership also unlocks:
