CoinDesk News: The Trump administration is about to finalize a voluntary framework that requires AI companies to submit their most advanced models to the government before releasing them to the public. About two weeks ago, the Office of the National Cyber Director at the White House distributed a draft of the framework to OpenAI, Anthropic, and Google, and the three companies jointly submitted their respective proposed revisions. An AI executive order signed by Trump in early June set August 1 as the deadline for creating the framework, aiming to establish a more standardized review process for powerful new AI that can quickly identify cyber vulnerabilities. As the deadline nears, some AI companies, during meetings with White House agencies, have expressed concerns about how the framework would define “frontier” models and whether closed-source and open-source models would be treated differently. The Trump administration insists that compliance with the framework is not mandatory.