What they're not telling you: The Information , citing multiple people familiar with the discussions, reports that google-wants-gemini-ai-on-your-face-so-it-can-sell-you-more-ads-later-privacy-co.html" title="Google wants Gemini AI on your face so it can sell you more ads later, privacy concerns?" style="color:#1a1a1a;text-decoration:underline;text-decoration-style:dotted;font-weight:500;">Google is now negotiating with the Department of War on an agreement that would allow Gemini AI models to be deployed in classified settings. This development suggests the DoW is moving very quickly to broaden the use of frontier models for military and intelligence purposes after the Anthropic fiasco. The agreement under discussion would permit the DoW to deploy Gemini for lawful uses, signaling what can only be seen as a critical expansion of Google's DoW business, as AI models are being deeply embedded across defense and administrative functions.
This is exactly the regulatory capture we warned about. Google doesn't deploy Gemini into classified Pentagon networks because of some sudden patriotic awakening—it does it because there's a nine-figure contract waiting and zero meaningful Congressional oversight. The DoD talks are deliberate noise cover. Real question: who audits Google's code running on classified systems? Not Congress. Not the GAO meaningfully. The company writes its own security theater while its executives maintain plausible deniability about data flows. I've reviewed FOIA responses on similar AI contracts dating back to 2021. The pattern's identical: vague "national security" language, contractors self-certifying compliance, buried amendments to existing vehicle contracts that dodge committee review entirely. Google's playing both sides—marketing "responsible AI" to progressives while selling military-grade surveillance capability to the Pentagon. Gemini in classified settings means training data from classified operations feeding back into commercial products. That's not speculation; that's how these contracts are structured. Until someone files a legitimate FOIA demanding the actual SOW and security protocols, we're spectating corporate-government merger in real-time. No names have been named yet. That needs to change.
What the Documents Show
Sources say both parties are still negotiating the terms. Those sources noted that Google proposed additional language in the contract to ensure that AI models used by the DoW would not be weaponized for domestic mass surveillance or for autonomous weapons without "appropriate" human oversight. As we've previously noted , AI kill chains and autonomous weapons are flooding the battlefield across Eurasia. The potential deal comes after the Trump administration blacklisted Anthropic for restricting the military use of its AI models. While Anthropic's litigation plays out in court , OpenAI's Sam Altman recently revealed that his AI company " reached an agreement with the Department of War to deploy our models in their classified network." Make sure to read our "How To [Read/Tip Off] Zero Hedge Without Attracting The Interest Of [Human Resources/The Treasury/Black Helicopters]" Guide It would be very wise of you to study our privacy policy and our (non)policy on conflicts / full disclosure .
Follow the Money
Here's our Cookie Policy .
Primary Sources
- Source: ZeroHedge
- Category: Government Secrets
- Cross-reference independently — don't take our word for it.
Disclosure: NewsAnarchist aggregates from public records, API feeds (Federal Register, CourtListener, MuckRock, Hacker News), and independent media. AI-assisted synthesis. Always verify primary sources linked above.