Trump administration nears voluntary framework letting government review advanced AI models before release, OpenAI Anthropic Google submit edits
The Trump administration is close to finalizing a voluntary framework letting AI developers submit advanced models to the government for review before release, per Bloomberg Law. Firms could give the government 30 days of access first. OpenAI, Anthropic and Google $GOOGL submitte
The Trump administration is close to finalizing a voluntary framework that would let AI developers submit their most advanced models to the federal government for review before public release, according to Bloomberg Law and The Information. The framework flows from an executive order President Trump signed on June 2, which directs federal agencies to build an opt-in, prerelease review process for frontier AI systems. Under the process, a developer can ask the government to determine whether a model qualifies as a 'covered frontier model.' If it does, the developer can choose to give the federal government up to 30 days of access to the model before its planned release to other trusted partners, and to collaborate with the government on selecting those partners. Participation remains voluntary throughout: companies are not required to secure government approval before releasing a model. The White House Office of the National Cyber Director circulated a draft of the framework to OpenAI, Anthropic and Google $GOOGL about two weeks ago, and the three companies jointly submitted their own edits, according to The Information. How the finalized framework will treat open-source and open-weight models, which can't be recalled once released, remains an open question that will shape how much practical oversight the review process actually provides.