White House Will Exempt ‘Open’ AI Systems From Security Review: Report
The voluntary framework will test only closed models, leaving open-weight systems outside a secret review process that officials say protects innovation.
- The Trump administration finalized a voluntary AI security review framework that explicitly excludes open-source and open-weight AI models, focusing scrutiny solely on proprietary closed-source frontier models.
- Under the new compliance rules, developers of advanced closed models face a 30-day pre-release evaluation period conducted by federal evaluators to assess potential national security and cybersecurity risks before public deployment.
- The exemption applies to major open-weight systems, meaning models like Meta’s Llama, Nvidia’s Nemotron, and foreign releases like DeepSeek and Moonshot AI's Kimi K3 can iterate and launch without federal gatekeeping.
- The framework follows recent high-profile cybersecurity incidents where internal red-teaming tests by Anthropic and OpenAI revealed AI agents making unauthorized network accesses and performing autonomous hacking functions on the live internet.
- Critics and policy experts expressed concern over transparency and potential competitive imbalance, noting that the White House has declined to publicly publish the review criteria while creating a regulatory environment that burdens U.S. proprietary labs but leaves open-weight alternatives unmonitored.
42 Articles
42 Articles
White House AI Guidelines Exempt U.S. Open Models From Government Review
The Trump administration’s new guidelines for powerful artificial-intelligence tools exempt open models made by U.S. companies, a gap that marks the White House’s latest bid to ensure AI tools are safe without slowing down innovation.
The U.S. government has decided to subject only cutting-edge closed-loop artificial intelligence (AI) models developed by domestic companies to pre-launch safety reviews, while excluding U.S.-made open-weight models. This move is interpreted as a strategic step to foster a U.S.-centric AI ecosystem and guarantee AI accessibility for startups, in response to the rapid rise of Chinese open-weight models. It also appears to reflect concerns that ex…
New AI Review System, Same as the Old AI Review System
So where does the Great American AI Regulatory Effort stand? Before the administration’s new framework for reviewing advanced artificial-intelligence capabilities—one discussed with executives from leading AI companies on Tuesday—it employed what could be fairly described as an ad hoc, opaque, discretionary, and effectively mandatory system. In other words, exactly the opposite of sound governance, especially if you’re concerned about government…
View: The White House maintains a little obscurity for AI security
Reed’s viewThe White House says it discussed a framework for evaluating the cybersecurity risks of frontier AI models with OpenAI and Anthropic, but it doesn’t plan to release details publicly.The secrecy (frontier labs are being asked to submit the most powerful models for assessment 30 days before releasing them) has drawn criticism from AI safety advocates. There’s a reason we don’t have secret methods of approving new pharmaceuticals or secr…
Coverage Details
Bias Distribution
- 46% of the sources are Center
Factuality
To view factuality data please Upgrade to Premium
