Microsoft, Google and xAI grant US Commerce Department early access to AI models

Microsoft, Alphabet-owned Google and Elon Musk’s xAI have agreed to give the U.S. government early access to new artificial intelligence models before public release, in a move that allows security reviews ahead of deployment. The arrangement was announced by the Commerce Department’s Center for AI Standards and Innovation, which will use the access to assess capabilities and national security risks.

The agreement adds the three companies to a growing group of AI developers working with federal officials on pre-release review of frontier systems. The government will be able to study the models before they reach the public, with a focus on how the systems perform and what risks they may pose.

The initiative reflects a broader push by federal agencies to get closer to advanced AI development as the technology becomes more powerful and more widely deployed. It also gives the Commerce Department a formal role in examining models before launch rather than after they are already in use.

For Microsoft, Google and xAI, the deal creates a direct channel for federal evaluation at an earlier stage of development. For the government, it expands access to some of the most closely watched AI systems now being built by leading technology companies.

The agreement does not appear to alter the companies’ product pipelines or public release plans, but it does add another checkpoint for models that may raise safety or security concerns. It also underscores the growing overlap between AI development and national security oversight as regulators seek more visibility into frontier systems before they are broadly deployed.