Trump Signs Executive Order Creating Voluntary 30-Day Federal Review Process for Advanced AI Models

President Donald Trump on Tuesday signed an executive order that creates a voluntary federal framework for reviewing emerging artificial intelligence models deemed to be on the “frontier” of development. The order gives AI companies the option to provide the federal government with access to covered frontier models for up to 30 days before they plan to release them publicly, a structure the White House says is intended to promote secure innovation and strengthen the cybersecurity of critical infrastructure.

The order responds to growing concerns about national cybersecurity risks tied to rapidly evolving AI systems while deliberately preserving a light regulatory touch. Rather than imposing mandates, the directive sets up a voluntary mechanism designed to encourage cooperation between AI developers and federal national security and cybersecurity agencies, enabling government reviewers to identify and remediate potential software vulnerabilities at scale during a limited review window.

As part of the initiative, the White House has tasked appropriate federal agencies with forming an AI cybersecurity clearinghouse. The clearinghouse, to be developed in voluntary collaboration with the AI industry and operators of critical infrastructure, will coordinate and “deconflict” scanning for software vulnerabilities. The administration emphasized that companies are not required to use the clearinghouse; participation and sharing under the framework remain voluntary.

The executive order expressly states that it should not be interpreted as creating any mandatory governmental licensing, pre-clearance, or permitting requirement for developing, publishing, releasing, or distributing AI models. That provision underscores the administration’s stated goal of balancing national security safeguards with the need to avoid stifling innovation or hindering U.S. industry competitiveness.

This approach contrasts with an earlier draft measure President Trump rejected on May 21, which would have allowed for a 90-day review period. At the time, Trump said he “didn’t like certain aspects of it” and that he wanted to avoid measures that might impede U.S. leadership in AI. “We are leading China, we’re leading everybody, and I don’t want to do anything that is going to get in the way of that lead,” he said in explaining his reasons for postponing the prior proposal.

Industry response to the new executive order has been largely positive among major AI companies and advisers. Dina McCormick, president of Meta AI, wrote on X that Meta “appreciate[s] the Administration’s support for public-private sector collaboration and look[s] forward to continuing to work with the White House as it implements the President’s plan.” OpenAI CEO Sam Altman said the U.S. should “lead on AI by continuing to develop the very best models, making sure they’re safe, and getting cyber tools into the hands of trusted defenders,” adding that “The new EO gets the balance right.” David Sacks, chair of the President’s Council of Advisors on Science and Technology, also praised the shift from a 90-day to a 30-day review window, calling it “a game changer” because it allows AI labs to comply without delaying new model releases.

Not all reactions were favorable. Some critics argue that a voluntary framework does not go far enough to mitigate the national security and societal risks posed by increasingly capable AI systems. Democratic Representative Don Beyer of Virginia described the order as “underwhelming policy that mirrors the Trump administration’s broader pattern of creating a wild west environment for AI development.” His comments reflect broader concerns among some lawmakers and experts that voluntary measures may leave gaps in oversight and accountability.

By establishing a limited, voluntary pre-release review window and creating an AI cybersecurity clearinghouse, the administration aims to strike a balance between enabling continued technological leadership and addressing cybersecurity vulnerabilities. The policy highlights an ongoing tension in U.S. AI policy: how to support innovation and competitiveness—particularly relative to China—while ensuring that the rapid deployment of powerful AI systems does not outpace the government’s ability to protect critical infrastructure and national security interests. The effectiveness of the voluntary framework will depend on industry participation and how federal agencies operationalize the clearinghouse and review process in the months ahead.