Microsoft, Google to share upcoming AI models with US government
What's the story
Leading tech giants Microsoft, Google, and Elon Musk's xAI have agreed to share their upcoming artificial intelligence (AI) models with the US government. The move is part of a new deal aimed at assessing national security risks before these advanced technologies are made public. The Center for AI Standards and Innovation (CAISI) will carry out these evaluations under the agreement.
Evaluation process
CAISI to evaluate AI models
The CAISI, a unit of the US Department of Commerce, will evaluate these AI models before they are deployed. The agency will conduct research to assess their capabilities and potential security risks. This comes after the global stir caused by advanced AI systems such as Anthropic's Mythos and OpenAI's GPT-5.4-Cyber, which have been criticized for their potential use in cyberattacks.
Testing hub
Over 40 evaluations conducted so far
CAISI, the government's main hub for AI model testing, has already conducted over 40 evaluations. These include cutting-edge models that are not yet publicly available. The agency said developers often provide stripped-back versions of their models without safety guardrails so it can probe for national security risks. This method has been used in previous agreements with OpenAI and Anthropic, established under the Joe Biden administration.
Military collaboration
Pentagon's agreements with AI companies
Last week, the Pentagon announced agreements with seven AI companies to deploy their advanced capabilities on the Defense Department's classified networks. The move is part of an effort to expand the range of AI providers working across the military.