Trump signs order inviting vetting of top AI models over security risks

Donald Trump has signed an executive order introducing a voluntary federal framework for vetting advanced artificial intelligence systems for national security risks, marking a shift in the White House’s approach to oversight of frontier models.

The order, issued on June 2, creates a mechanism for the federal government to review the most advanced AI systems for up to 30 days before public release, Caliber.Az reports, citing an American news agency.

Participation by developers is voluntary, according to the text, and is aimed at identifying potential national security vulnerabilities in cutting-edge systems.

“Advanced AI capabilities make our Nation stronger, but also introduce new national security considerations that require coordinated action across executive departments and agencies,” the order says.

The policy comes less than two weeks after Trump declined to sign an earlier version of the directive, citing concerns it could slow US technological progress. The administration has framed the revised version as more limited in scope than initially proposed, focusing only on frontier models rather than broad-based AI regulation.

In a social media post, the White House said the executive order "creates a process for frontier labs to voluntarily share cutting-edge cyber models in order to secure critical infrastructure and strengthen the government’s own cyber defences. We are NOT conducting oversight of all new models, as that level of government overreach would have chilling effects on free speech and innovation."

Industry response was broadly supportive. Anthropic described the move as a positive step and said it looks forward to cooperating with implementation, while OpenAI and Google also welcomed the framework, emphasising the importance of structured safety systems for increasingly capable models.

Juan Londoño, a policy analyst at the libertarian-leaning Cato Institute, said the order was “a step in the right direction to prepare the nation for the release of advanced AI systems.”

He added concerns about governance discretion, warning that the mechanism could be inconsistently applied. Londoño said in an interview that giving significant authority to the director of the National Security Agency represented a “dangerous precedent” that could enable the government to “weaponize” the policy against firms it is in dispute with.

The order follows heightened scrutiny of advanced AI systems after Anthropic’s release of a new model and growing concern inside government and financial institutions about cybersecurity implications. Officials including Treasury Secretary Scott Bessent and Federal Reserve Chair Jerome Powell have reportedly raised alarms in meetings with Wall Street executives about potential vulnerabilities exposed by frontier systems.

Anthropic said the policy is “an important step in strengthening America’s leadership in AI” and signalled support for continued collaboration with the White House.

By Aghakazim Guliyev