Microsoft, Google, xAI grant US government early AI access, citing security

Microsoft, Google, and Elon Musk’s xAI will provide the U.S. government early access to new artificial intelligence models.

The Department of Commerce’s Center for AI Standards and Innovation will evaluate model capabilities and security risks before public release.

This collaboration aims to identify potential threats such as cyberattacks and military misuse.

Microsoft will work with government scientists to test for unexpected behaviors.

The company will also develop shared testing datasets and procedures with the government.

Microsoft maintains a similar security agreement with the UK’s AI Security Institute.