Google, Microsoft, xAI sign US safety pacts, vetting models before release

The U.S. Department of Commerce’s National Institute of Standards and Technology (NIST) signed agreements with Google DeepMind, Microsoft, and xAI. These partnerships allow the Center for AI Standards and Innovation (CAISI) to evaluate frontier AI models before public release.

CAISI will assess advanced capabilities and security risks to establish federal safety and governance standards. The research focuses on unreleased models to identify national security implications and develop commercial best practices.

CAISI has already completed more than 40 evaluations of various AI models. This initiative formalizes public-private cooperation to regulate the future of the artificial intelligence sector.