Hundreds of Google employees urge company to block Pentagon AI deal

More than 600 Google employees signed a letter to chief executive Sundar Pichai urging the company to block the Pentagon from using Google’s AI models for classified work, intensifying an internal revolt over whether artificial intelligence should be allowed into military secrecy. Organizers said many signers work in Google DeepMind, and more than 20 are principals, directors or vice presidents, a sign that concern has spread beyond junior staff.

The employees said they were “deeply concerned” about ongoing negotiations between Google and the U.S. Department of Defense, and pressed the company to refuse any arrangement that would put its systems into classified military settings. The letter also called for more transparency around existing Pentagon contracts and demanded an ethics board with employee representation, reflecting a broader push to give workers a formal role in decisions that could shape how the company’s technology is used in war and surveillance.

The clash centers on reported talks over a deal that would let the Pentagon use Google’s Gemini models in classified environments. Reuters reported on April 16, 2026, that the parties were discussing use of Google AI for “all lawful uses,” including classified workloads. Google already has a separate Defense Department contract for non-classified work through a platform known as genAI.mil, and the new proposal would extend that relationship into classified domains.

For critics inside the company, that shift raises more than a business dispute. Employees warned about autonomous weapons and mass surveillance, arguing that AI systems can centralize power and make mistakes in ways that are harder to detect once they are embedded in secret military programs. The issue has also revived memories of Google’s 2018 Project Maven fight, when worker pressure helped push the company to abandon a Pentagon AI program tied to drone operations.

The pressure on Google comes as major AI firms face growing scrutiny over defense contracts and military limits. Only weeks earlier, Anthropic was reportedly dropped by the Defense Department after seeking similar restrictions on military use of its AI models. For Google, the fight over Gemini is now about more than one contract. It has become a test of whether executives, workers or governments will decide the ethical boundaries of AI inside one of the world’s most powerful technology companies.