OpenAI

U.S. Will Test Frontier AI Before Release as Security Reviews Expand

The pre-release checks signal a shift toward government vetting after concerns about Anthropic’s Mythos model.