President Donald Trump on Tuesday signed a new executive order establishing a voluntary process for some artificial intelligence companies to share their most advanced AI models with the federal government before public release, marking a narrower approach to oversight after the White House abandoned a stricter proposal last month.
The order is intended to address growing concerns that increasingly powerful AI systems could pose cybersecurity and national security risks while avoiding regulations that administration officials say could hinder U.S. competitiveness in the global AI race, particularly against China. The White House said the order seeks to balance innovation and security. It directs agencies including the Departments of Defense, Homeland Security and Treasury, along with the National Security Agency, to strengthen AI-related cybersecurity efforts and establish mechanisms for identifying software vulnerabilities that advanced AI systems could uncover or exploit.
Under the new guidelines, developers of certain “frontier” AI models may voluntarily provide the government with access to those systems for up to 30 days before they are released to trusted partners or the public. Federal agencies would use that period to assess potential cybersecurity vulnerabilities and evaluate risks associated with advanced AI capabilities.
The administration also emphasized that participation is optional and that the order does not create any licensing, permitting or mandatory pre-approval requirements for AI developers. The directive states: “Nothing in this section shall be construed to authorize the creation of a mandatory governmental licensing, preclearance, or permitting requirement for the development, publication, release, or distribution of new AI models, including frontier models.”
The measure follows weeks of internal debate within the administration and lobbying from technology industry leaders. The policy debate intensified after AI developer Anthropic disclosed the capabilities of its experimental Mythos model, which demonstrated an ability to identify security weaknesses in computer systems. The company initially limited access to the technology while selected organizations assessed potential risks and strengthened defenses.
Editorial credit: IAB Studio / Shutterstock.com