world
US Government to Test New AI Models From Google, Microsoft and xAI Before Release
The US Department of Commerce announced agreements with Google, Microsoft and xAI to test new AI models for capabilities and security risks before public release. The pacts expand on prior arrangements with OpenAI and Anthropic, with evaluations focusing on national security, cybersecurity and biosecurity. CAISI has already completed over 40 evaluations of unreleased models.
SE
New York PostAudio version
Tap play to generate a narrated version.
The US Department of Commerce announced that Google, Microsoft and xAI have agreed to submit their new AI models for testing before public release. The agreements were reached through the Center for AI Standards and Innovation (CAISI), according to the department.
CAISI has conducted more than 40 evaluations of AI tools, including state-of-the-art models that remain unreleased, the center stated. The evaluations will cover testing, collaborative research and best practice development for commercial AI systems. The pacts give the US government pre-release access to the most advanced frontier AI models from the companies.
The agreements build on prior voluntary arrangements with OpenAI and Anthropic. Those earlier agreements, reached in 2024 under the Biden administration when the entity was known as the US Artificial Intelligence Safety Institute, have been renegotiated to align with new directives. The AI Safety Institute was established in 2023 and renamed CAISI under the Trump administration.
The Trump administration pledged in July to partner with technology companies to vet AI models for national security risks. Models from Google, Microsoft, xAI, OpenAI and Anthropic will undergo evaluations for national security, cybersecurity and biosecurity risks, according to announcements from the department.
Tom Lue, vice president of global AI affairs at Google DeepMind, confirmed the partnership in a social media post. The announcement was made on 5 May 2026. It coincided with a separate Pentagon agreement with seven tech companies—Google, Microsoft, Amazon Web Services, Nvidia, OpenAI, Reflection and SpaceX—to use AI in classified systems.
Anthropic developed a model called Mythos. The company is involved in a lawsuit with the US Department of Defense over its refusal to drop safety guardrails for government use of its models. Senior members of Trump’s staff met last month with Anthropic CEO Dario Amodei. President Donald Trump signed executive orders last year that formed the basis of his administration’s AI Action Plan.
