Google’s recent agreement to provide its Gemini AI models to the Pentagon for “any lawful purpose” has triggered employee backlash reminiscent of the 2018 Project Maven controversy, but this time employee leverage appears significantly diminished. In 2018, when Google partnered with the Pentagon on Project Maven (an initiative using AI to analyze drone surveillance footage for targeting workflows) employee protests forced the company to abandon the contract and establish AI principles pledging not to develop AI for weapons or surveillance violating international norms. However, Google updated these principles in February 2025, removing the explicit pledge from its public website.
The current deal follows similar agreements by OpenAI and xAI, allowing AI models to be deployed in classified military networks. Close to 600 employees initially signed an open letter opposing the contract, though this number has grown to around a thousand signatures. Unlike the Project Maven situation, Google’s leadership has maintained a defiant stance, stating in an internal memo that the company “proudly” works with the U.S. military and intends to continue.
Google DeepMind research scientist Alex Turner publicly criticized the deal, stating that “Google affirms it can’t veto usage, commits to modify safety filters at government request, and aspirational language with no legal restrictions”. Former Google employee Laura Nolan, who resigned over Project Maven, explains that employee influence has eroded due to cost-cutting, layoffs across the tech sector, and Google’s crackdown on internal communication platforms following the Maven incident.
Legal experts, including Charlie Bullock from LawAI’s U.S. Law and Policy team, argue that Google’s contract language appears “less restrictive and more permissive” than OpenAI’s deal, particularly regarding mass surveillance protections and the obligation to remove technical safeguards at government request. Critics worry about precedent-setting for future AI systems regarding autonomous weapons and mass surveillance capabilities.
Reference
Nolan, B. (2026, May 4). A new Google AI deal with the Pentagon has sparked employee backlash. Their leverage appears limited. Fortune. https://fortune.com/2026/05/04/google-employee-backlash-pentagon-ai-contract-power-waned-since-project-maven/