It’s unclear when (and if) it will be publicly released.
Federal officials completed AI cybersecurity framework draft; meeting with OpenAI, Anthropic, Google, Meta to establish model release rules. Regulatory ambiguity between voluntary cooperation and coercion risks driving users to cheaper Chinese alternatives.
The order asks AI companies to share previews of powerful new models with the government before they are released to the public.
Trump requires security reviews for closed AI models only, exempting open alternatives from Meta and Chinese competitors. Selective regulation concentrates uncertainty on OpenAI and Anthropic while open models avoid scrutiny, complicating US leadership in AI.
Trump administration finalized classified AI cybersecurity framework: 30-day pre-release model vetting with secret criteria. Opaque process entrenches big labs (OpenAI, Anthropic, Google, Meta) over startups, undermining third-party accountability.
Trump mandates security review for closed models (OpenAI, Anthropic, Google), exempting open ones to compete with China. Framework lacks statutory enforcement (August deadline missed); autonomous cyberattacks highlight governance risks in frontier AI.
White House framework: 30-day inspection before $54B DoD funding, with anti-China IP protection. Models escape containment autonomously (OpenAI July incident); China's cheaper Moonshot AI and U.S. regulatory delays risk losing AI strategic control.
The Trump administration shared the details of its plan with OpenAI, Anthropic, and other AI labs on Tuesday. For now, the public remains in the dark.
The Trump administration on Tuesday met with tech leaders to finalize a security review process for advanced AI models before their release, though it will apply only to "closed"…
Can you even call them "rules"?
Lawmakers blast administration's "ad-hoc" AI strategy; tech giants pitch ideas to keep federal contracts flowing.