Google eyes AI deal with Pentagon to deploy Gemini AI in classified ops

Google is reportedly engaged in discussions with the U.S. Department of Defense regarding the potential deployment of its advanced Gemini AI models in classified military operations. According to a report by The Information, this agreement is still in negotiation, and if finalized, it could allow the Pentagon to leverage Google’s AI technology within legal and regulatory parameters. The proposed contract includes specific provisions aimed at regulating the application of the AI model, ensuring it is not used for controversial purposes. These guidelines are designed to prevent the technology’s application in domestic mass surveillance or in autonomous weapons systems without appropriate human oversight. A Pentagon representative has confirmed the agency's ongoing collaborations with tech companies to integrate advanced AI technologies, although they did not specifically confirm the discussions with Google. This potential partnership marks a significant opportunity for Google to deepen its relationship with the U.S. government. In a separate context, OpenAI had previously established an “All Lawful Purposes” agreement with the Pentagon in February 2026, allowing the use of its AI models for classified tasks. This agreement faced backlash from OpenAI users concerned about the implications for mass surveillance, prompting the company to clarify that its AI tools are not intended for such uses and are restricted to cloud-based operations. Meanwhile, Anthropic's relationship with the Pentagon has become more complicated, with the company facing a possible ban due to perceived supply-chain risks. Anthropic is currently pursuing legal action against the government, claiming that the actions taken against it are unlawful retaliation for its commitment to public safety.