President Donald Trump signed an executive order Tuesday establishing a voluntary system for the federal government to review powerful artificial intelligence models before tech companies release them to the public.
Under the framework, companies would be asked to submit their AI systems to government reviewers up to 30 days before launch. The Trump administration says the process will strengthen national security by identifying potential cybersecurity threats before new models reach consumers.
The order stops short of requiring companies to participate. Tech firms can choose whether to undergo federal review, marking a compromise between hardline voices pushing for mandatory vetting and industry advocates pressing for minimal regulation.
The move signals a striking shift in Trump's AI posture. One of his first presidential acts was scrapping a Biden-era executive order that set safety standards for AI development. Now, just weeks later, he is embracing government oversight of the sector.
The timing reflects growing anxiety over advanced AI capabilities. Anthropic's latest model has triggered alarms among safety experts and government officials for its sophisticated ability to identify and exploit software vulnerabilities at scale.
Trump's administration has already been moving behind the scenes on this front. Last month, officials announced preliminary agreements with Microsoft, Google DeepMind and xAI to review their emerging AI systems. The Commerce Department's Center for AI Standards and Innovation has similar arrangements with OpenAI and Anthropic, though the federal government recently removed details of these deals from its website without explanation.
The National Security Agency and Department of Defense will identify which AI models require government review. The Treasury Department will take the lead in stress-testing models to find weaknesses. Officials are also pushing to hire more cybersecurity and AI specialists across federal agencies and to strengthen defenses at critical infrastructure including hospitals and financial institutions.
The executive order represents one piece of Trump's broader AI strategy. In December, he announced a separate order aimed at blocking states from creating their own AI regulations, establishing a federal taskforce to challenge state-level AI laws.
Free speech advocates have cautioned that expanded government involvement in AI vetting could create opening for censorship, though the administration frames the review process as standard national security practice.
Author James Rodriguez: "Trump's sudden embrace of AI oversight is a jarring reversal that proves his deregulatory instincts bend when national security concerns grab his attention."
Comments