Microsoft, Google, and Elon Musk's xAI Just Handed the US Government a Massive AI Power

Three of the world’s most powerful tech companies just made a shocking move: Microsoft, Google, and Elon Musk’s xAI are handing over their cutting-edge artificial intelligence models to the US federal government for national security testing.

The Center for AI Standards and Innovation (CAISI) at the Department of Commerce announced the deal on Tuesday, and the timing is explosive. It comes amid growing panic in Washington about Anthropic’s newly unveiled Mythos AI model and how it could be weaponized by hackers.

Here’s what’s actually happening: The US government will now get early access to evaluate these models before they hit the market, conducting rigorous assessments to identify security vulnerabilities and potential national security threats. Think cyberattacks. Think military misuse. Think worst-case scenarios.

Microsoft is taking this seriously. The company says it will work directly with US government scientists to test AI systems “in ways that probe unexpected behaviors.” They’re building shared datasets and testing workflows specifically designed to break things. Google and xAI are following suit, though Google declined to comment and xAI hasn’t responded yet.

“Independent, rigorous measurement science is essential to understanding frontier AI and its national security implications,” CAISI Director Chris Fall said. Translation: We need to know exactly what these things can do before bad actors get their hands on them.

This isn’t new territory. CAISI has already completed over 40 evaluations on cutting-edge models that aren’t even public yet. Developers regularly strip away safety guardrails on their AI versions so government researchers can hunt for vulnerabilities. It’s like handing over a brand-new car to mechanics to find every possible way it could crash.

The move fulfills a pledge President Trump made in July to partner with tech companies and vet their AI models for national security risks. It builds on earlier agreements from the Biden administration with OpenAI and Anthropic.

But here’s the kicker: Anthropic, the company behind the alarming Mythos model that sparked all this concern, is noticeably absent from major Pentagon agreements. The AI company is locked in a public dispute and legal battle with the Trump administration over the ethics and safety of using AI in warfare.

Meanwhile, Wall Street’s reaction was mixed. Microsoft’s stock dipped 0.6 percent, while Alphabet (Google’s parent) climbed 1.3 percent. xAI remains private.