The Trump administration issued an executive order directing federal agencies to establish a pre-release review process for frontier AI models. Companies must submit models to government reviewers before making them available to the public, with the goal of assessing national security and safety risks.
The order defines 'frontier' models as those with computational capability above a specified threshold and applies to major AI labs including OpenAI, Anthropic, Google, and Meta. Implementation details remain unclear, but the directive aims for a 30-day review window per submission.
AI industry leaders expressed mixed reactions. Some support regulatory clarity; others warn that the requirement could slow innovation and push development offshore. Experts debate whether executive authority extends to premarket AI review or if Congress must authorize such a regime.
