AI news ·
Microsoft, Google and xAI agree to give US government early access to AI models for security testing
The Trump administration is requiring Microsoft, Google, and xAI to submit new AI models for government security testing before public release. The Department of Commerce will assess capabilities and risks.
Trump Administration to Test AI Models Before Public Release
The Trump administration is requiring Microsoft, Google, and Elon Musk's xAI to provide early access to new artificial intelligence models for government security testing before deployment, according to a pact announced this week.
The Centre for AI Standards and Innovation at the Department of Commerce will evaluate the models and assess their capabilities and security risks. The agreement fulfills a pledge the administration made in July 2025 to vet AI systems for national security threats.
The move reflects growing concern in Washington about powerful AI systems. Officials want to identify threats ranging from cyberattacks to military misuse before tools reach the public.
How the Testing Will Work
Microsoft said it will work with US government scientists to test AI systems "in ways that probe unexpected behaviours." The company and government will develop shared datasets and workflows for evaluation.
Microsoft has signed a similar agreement with the UK's AI Security Institute, the company said in a statement.
A Shift in Trump's AI Approach
The early-access requirement marks a change in direction for President Donald Trump, who has previously championed AI and sought to ease regulatory burdens to support data centre construction and power supply.
The White House briefed executives from Anthropic, Google, and OpenAI on the oversight plans last week, according to reporting on the proposed executive order. The administration is considering creating a working group on AI to strengthen oversight of the technology.
One proposal includes a formal government review process for new models before release.
What This Means for Your Organization
If your company develops or deploys AI models, expect government review timelines to lengthen. Pre-release testing requirements will affect product roadmaps and go-to-market strategies.
Understanding these regulatory expectations now helps teams plan ahead. For management guidance on AI implementation and policy, see resources on AI for Executives & Strategy and AI for Management.