White House discusses executive order for government review of new AI models before release, prompted by Anthropic's Mythos model

The White House is considering an executive order establishing a working group of tech executives and government officials to examine oversight procedures, including formal government review of new AI models before public release. The White House briefed representatives from Anthropic, Google, and OpenAI on these plans last week.

Anthropic's Claude Mythos model—which the company says is powerful enough at identifying software vulnerabilities that it declined to release publicly—appears to have triggered the policy shift. The NSA is already using Mythos to assess vulnerabilities in US government software.

This reversal marks a sharp change from Trump's earlier deregulation agenda. In July, Trump called AI a 'beautiful baby' that should not be stopped with 'foolish rules' or 'even stupid rules,' and Vice President JD Vance warned that 'excessive regulation' could 'kill a transformative industry.' Within the White House, concern has reportedly grown over potential political fallout from an AI-enabled cyberattack.