The Trump administration is weighing an executive order to create an AI working group that would implement a formal government review process for new, powerful AI models, according to reports from platformer.news.
U.S. officials, speaking on condition of anonymity to discuss sensitive policy deliberations, told the New York Times that the group would bring together tech executives and government officials to examine oversight procedures. The proposed review process could mirror a system currently being developed in Britain, which uses government bodies to ensure AI models meet specific safety standards.
This shift marks a reversal of the administration's recent stance on AI regulation. In February 2025, Vice President JD Vance warned at the Paris AI Action Summit that excessive regulations might "kill a transformative industry just as it’s taking off." Vance also suggested that companies seeking regulation might be attempting to crush future competitors.
White House AI and crypto czar David Sacks previously characterized fears regarding AI capabilities as a "doomer industrial complex" using a "sophisticated regulatory capture strategy based on fearmongering." Similarly, Michael Kratsios, head of the Office of Science and Technology Policy, has criticized international governance efforts for maintaining a "general atmosphere of fear."
The impact of Anthropic's Mythos
The administration’s change in direction appears driven by the capabilities of Anthropic’s latest large language model, Mythos. According to platformer.news, the model has demonstrated an ability to develop cybersecurity exploits that the government views as a national security risk.
The White House is currently opposing Anthropic’s plan to expand access to Mythos from roughly 50 companies to 120. While the administration also claims concerns regarding Anthropic's available compute power, the company has denied those claims.
The situation is further complicated by the administration's simultaneous efforts to designate Anthropic as a "supply chain risk." This designation stems from the company's refusal to amend a Pentagon contract to allow for "all lawful use" of its technology. Despite this, some White House officials are working to expand agency access to the model, even as others work to phase out Anthropic's technology over the next six months.
Industry leaders are already responding to the shifting regulatory landscape. Google, Microsoft, and xAI announced on Tuesday that they will provide the U.S. Commerce Department’s Center for AI Standards and Innovation with early access to their models for review.
This move follows the administration's recent restructuring of the U.S. AI Safety Institute into the Center for AI Standards and Innovation. Commerce Secretary Howard Lutnick previously stated that innovators would no longer be limited by the standards used under previous administrations.
