US government now has pre-release access to AI models from five major labs for national security testing

The US Department of Commerce's Center for AI Standards and Innovation (CAISI) has signed new agreements with Google Deepmind, Microsoft, and xAI to test advanced AI models for national security risks before public release. Previously, CAISI had similar agreements with Anthropic and OpenAI. Testing occurs in classified environments, including evaluations of unreleased models and versions with reduced safety guardrails. This expansion comes as AI models rapidly improve at exploiting security vulnerabilities and the tech race with China intensifies.