Over 580 Google employees, including 20+ directors and senior DeepMind researchers, signed a letter urging CEO Sundar Pichai to reject classified military AI work for the Pentagon. The core concern is that on air-gapped classified networks, Google loses all ability to monitor or limit how its AI tools are used, leaving only Pentagon assurances as a guardrail against autonomous weapons or mass surveillance. The letter echoes the 2018 Project Maven protest, but the stakes are far higher: Google has since removed weapons language from its AI principles, deployed Gemini to 3 million Pentagon personnel, and is now negotiating 'all lawful uses' classified access. The Pentagon's AI budget has surged to $54.6 billion for FY2027, and the administration has shown willingness to blacklist companies like Anthropic that resist unconstrained military use. Despite notable seniority among signatories, the gap between internal dissent and corporate decision-making has widened significantly since 2018.

8m read timeFrom thenextweb.com
Post cover image
Table of contents
The historyThe negotiationThe stakesThe gap
1 Comment

Sort: