Google has announced it will now allow users to remove visible watermarks from its AI-generated images, videos and songs. While the invisible SynthID watermark and C2PA metadata will remain, users can opt out of the visible ones in settings.

This change reflects an evolving approach to labelling AI-generated media, as visible watermarks often hinder professional use but are necessary for transparency. Google’s VP for Gemini, Josh Woodward, stated that this toggle will be available for Nano Banana, Omni and Lyria models, with support rolling out soon in Gemini, Flow video editor, and Search.

The move comes after Anthropic’s decision to include watermarks in Claude-generated content to comply with EU regulations. Google is also open sourcing a library called Credentio to allow developers to embed local validation mechanisms in their apps.

With this feature, users will have more creative control over their AI-generated content but may face challenges in identifying and attributing such media. The balance between privacy and transparency remains a delicate line to walk.