Google Makes Visible Gemini Watermarks Optional for AI Images, Videos and Music

Google is making a change to its Gemini AI platform, allowing users to disable a visible watermark that identifies AI-generated images, videos, and music. The option will be available in the platform's Settings under Media Watermark. This feature applies to content created with Google's Nano Banana image model, Omni video model, and Lyria music model.

The invisible SynthID watermark and C2PA Content Credentials will still be embedded in the content, providing transparency about its origin. The visible watermark, which consists of a small sparkle icon, will disappear once users toggle it off. Google's VP Josh Woodward explained that the company is balancing creative control with safety, while still maintaining systems for transparency.

Users will still be able to determine if an image was AI-generated using Gemini or Search, even if the visible watermark is disabled. This move comes as some jurisdictions are considering mandatory visible labeling for AI-generated content, while other AI companies are relying more on invisible markers and metadata. Google's approach is to prioritize behind-the-scenes verification, which may require users to take an extra step to verify AI-generated content.