Google faces privacy class action over Gemini access to Gmail, Chat and Meet

What happened

A proposed class action in the U.S. alleges that Google enabled Gemini inside Gmail, Chat and Meet accounts without user consent, and that Gemini accessed, reviewed or analyzed users' communications. The complaint alleges on information and belief that scanned email content may be used to train Google's higher-tier AI models and may be seen by human reviewers.

Details

• The proposed class covers U.S. residents with Google accounts whose Gmail, Chat or Meet communications were accessed by Gemini after the feature was enabled without consent.
• The second amended complaint adds a claim under California's Unfair Competition Law and recasts the Stored Communications Act claim around the bar on providers divulging stored communications.
• On September 29, 2026, Google filed a motion to dismiss the second amended complaint. The case is one of the first major tests of whether wiretapping and communications-privacy laws reach an AI assistant built into everyday email and messaging.

Why it matters

If email content processed by an AI assistant is fed back into model training, users' own writing and private correspondence become training data without a license or consent.