Trump Signs Executive Order for Voluntary Government Review of AI Models Before Release
President Donald Trump signed an executive order Tuesday establishing a voluntary framework allowing the federal government to review powerful new artificial intelligence models up to 30 days before they are publicly released — a move that balances national security concerns against pressure from the tech industry to avoid heavy regulation.
What the Order Does
Under the new guidelines, AI developers may voluntarily submit their models to federal review for a period of up to 30 days prior to public release. Companies would first engage the government to assess how risky their model may be and whether it would benefit from the review process.
The National Security Agency and the Department of Defense will help determine which AI models require government scrutiny, while the Treasury Department will play a key role in identifying vulnerabilities. The order also directs the government to hire more cybersecurity and AI professionals and strengthen cybersecurity systems at critical institutions, including rural hospitals, community banks and local utilities.
What the Order Does Not Do
Crucially, the order stops well short of imposing mandatory review requirements on tech companies. The text explicitly states that nothing in the order “shall be construed to authorize the creation of a mandatory governmental licensing, preclearance, or permitting requirement” for developing or releasing new AI models.
The order also declares that the United States “refuses to stifle this innovation with overly burdensome regulation” — language that signals Trump is prioritizing growth over oversight, consistent with his broader deregulatory posture toward the tech sector.
A Scaled-Back Version of an Earlier Plan
The signed order is considerably weaker than earlier drafts. An initial version of the order reportedly required companies to submit models 90 days before release and included mandatory review provisions — features that drew fierce opposition from Silicon Valley.
Trump postponed the original order in late May, citing concerns that it was too restrictive. Reports from multiple news outlets indicated that tech billionaires including Elon Musk, Mark Zuckerberg and former White House “AI czar” David Sacks personally lobbied the president by phone to reverse course.
“I didn’t like certain aspects of it, I postponed it,” Trump said in the Oval Office on May 21. “We’re leading China, we’re leading everybody, and I don’t want to do anything that’s gonna get in the way of that lead.”
Existing Agreements Already in Place
The executive order formalizes and expands arrangements already underway. Last month, the Trump administration reached agreements with Microsoft, Google DeepMind and xAI to review early versions of their AI models before release, though the federal government recently removed details of that agreement from its website without explanation.
The Center for AI Standards and Innovation (CAISI), part of the Department of Commerce, already holds similar agreements with OpenAI and Anthropic. Officials describe such information-sharing arrangements as standard practice and essential to national security, though some free speech advocates have warned that expanded government access could create conditions for censorship.
The Security Backdrop
The order comes amid growing concern about the potential dangers of cutting-edge AI systems. Anthropic’s Mythos, a model with advanced cybersecurity capabilities, has alarmed AI safety experts, governments and tech companies alike for its reported ability to exploit vulnerabilities in widely used software at unprecedented scale.
Trump signed a separate AI-focused executive order in December aimed at preventing individual states from regulating AI, establishing a federal task force to challenge state-level AI laws — a move that has drawn scrutiny from those who argue it undermines legitimate regional governance.
