You can now turn off Google Gemini’s visible watermarks

Details

Google updated Gemini and its AI video generator Flow to allow users to remove the visible 'sparkle' watermark from content generated by the Nano Banana and Omni models. The toggle is in Gemini's Settings under 'Media Watermark.' Despite the visible mark becoming optional, AI-generated content retains invisible SynthID watermarks and C2PA metadata embedded in the file, allowing Gemini and Google Search to confirm AI origin when queried. Google plans to extend the toggle to Search, though it will not be available in countries legally requiring visible AI watermarks. Josh Woodward, VP of Google Labs, Gemini, and AI Studio, framed the change as balancing creative control with safety. The shift aligns Google with what other major AI image generators have already done — OpenAI uses SynthID and C2PA without visible marks, Meta launched its Content Seal standard, and Anthropic announced this week it is applying invisible watermarks to Claude-generated text and images for EU compliance. The practical consequence: AI-generated content becomes visually indistinguishable from human-created work, with provenance tracked only through metadata standards most consumers don't know how to access.