Google lets users remove visible AI watermarks from creations

Google will allow users to remove the visible watermark from content generated with its AI models, including images, videos, and songs. The change, announced by Josh Woodward, VP of Gemini, applies to models Nano Banana, Omni, and Lyria, and will roll out to Gemini and Flow first, with Search to follow. Users can toggle the visible mark via Settings > Media Watermark. However, invisible identification remains mandatory: Google will keep SynthID and C2PA metadata to trace AI-generated content. Woodward said the move balances creative control and security, addressing concerns that visible watermarks interfere with professional and creative uses. The company also open-sourced Credentio, a library for provenance. Source: lavozar Source: infobae