GOOGLE LETS USERS REMOVE VISIBLE WATERMARKS FROM AI CONTENT

Google will allow users to remove visible watermarks from AI-generated images, videos and music, according to an announcement by Josh Woodward, vice president of Google Labs, Gemini and AI Studio. A toggle for a "Media watermark" setting will become available in Gemini and Google's video editor Flow, with support for Search coming soon. The toggle will apply to content created using Google's Nano Banana, Omni and Lyria models, which currently display a distinctive sparkle icon to denote AI generation. The rollout will occur everywhere except countries where visible watermarks are required by law.

The EU and South Korea have both mandated visible watermarks on AI-generated content, particularly deepfakes that appear authentic. Google's change means the visible sparkle watermark can be toggled off, though invisible watermarking systems will remain. Woodward stated that invisible SynthID watermarks and C2PA metadata will continue to be embedded in all AI-generated content and cannot be disabled by users. Those seeking to verify whether content is AI-generated can ask Gemini or Google Search to analyse it, though this requires active user intervention rather than instant visual identification. Woodward described the decision as "striking a balance here between creative control and safety," noting that visible watermarks can hinder creative professionals whilst protecting against undetected deepfakes.

The move reflects broader industry shifts in AI labelling practices. OpenAI and Meta do not apply visible watermarks to their AI generators, instead relying on invisible tracking methods. Anthropic recently announced it was adding C2PA-based invisible watermarks to text and files generated by Claude to comply with EU regulations. Music platform Suno has introduced transparency tools showing whether songs are AI-generated when shared across platforms.