White House considers government vetting of AI models before public release in major policy reversal

The Trump administration is considering an executive order that would establish a working group of tech executives and government officials to create a formal government review process for new AI models before public release, the New York Times reported on May 4. White House officials briefed executives from Anthropic, Google, and OpenAI on the plans during meetings last week. The proposal marks a sharp reversal from Trump's hands-off approach: on his first day in office in 2025, he revoked a Biden order requiring AI developers to share safety test results before release. A White House official characterized discussion of a potential EO as speculation, but the reporting was corroborated by Bloomberg, Forbes, and Reuters. The review process could be modeled on a UK framework that assigned multiple government agencies to evaluate AI safety standards. The policy shift follows the Pentagon's decision to exclude Anthropic from classified AI contracts over safety disputes.