Back to all articles
•Technology

Google Introduces Watermark Control in AI Content with New Features

View original source

Google has announced a significant update allowing users to remove visible watermarks from AI-generated content such as images, videos, and songs. Key details include:

  • This change does not impact the invisible SynthID watermark, used for transparency and tracking, or the C2PA standard-related metadata.
  • Josh Woodward, Google's VP for Gemini, made the announcement on X, explaining the feature will be available on the Nano Banana, Omni, and Lyria models.
  • The toggle feature to remove visible watermarks will be integrated into Google's Gemini and Flow video editor, with future support for Search.
  • Google's approach balances creative control with safety; while users can disable visible watermarks, invisible tracking remains for transparency.
  • The update aims to make AI-generated content more usable for professional applications without affecting content identification.
  • Google is also open sourcing a library called Credentio, allowing developers to embed local validation mechanisms into their apps.
  • This move follows Anthropic's controversial decision to add watermarks to AI-generated text in compliance with EU regulations.