Event
Google has updated its Gemini and Flow AI tools to allow users to toggle off visible watermarks, such as the 'sparkle' logo, from generated images, videos, and music. While visible markers are now optional, Google will continue to embed invisible SynthID and C2PA watermarks and metadata to verify AI origin. The feature will not launch in countries where visible watermarks are legally required.
2 published stories · artificial-intelligence · source reliability 67%
Users can toggle off the 'sparkle' logo from images, videos, and music, though invisible SynthID and C2PA metadata will remain embedded to verify origin.
Background
Google has updated its Gemini and Flow AI tools to allow users to disable visible watermarks, such as the 'sparkle' logo, from generated images, videos, and music. This change grants users greater creative control, aligning Google’s approach with competitors like OpenAI and Meta, who rely on invisible markers rather than visible overlays. Despite the removal of visible indicators, Google will continue to embed invisible SynthID watermarks and C2PA metadata to verify the origin of AI-generated content. The toggle feature will not be available in jurisdictions where visible watermarks are legally mandated. This update follows Gemini surpassing one billion monthly active users, placing it on par with OpenAI’s ChatGPT.
Google has updated its Gemini and Flow AI tools to allow users to disable visible watermarks, such as the 'sparkle' logo, from generated images, videos, and music. This change grants users greater creative control, aligning Google’s approach with competitors like OpenAI and Meta,
Fact cards
Key entities: Google, Gemini, Flow, SynthID, C2PA, Josh Woodward, OpenAI, Meta.
What can users now do regarding watermarks in Gemini and Flow? Users can toggle off visible watermarks, such as the 'sparkle' logo, from generated images, videos, and music. Will AI-generated content still be identifiable as AI? Yes, Google will continue to embed invisible SynthI
The Verge | https://www.theverge.com/tech/980416/google-gemini-ai-watermarks-removal | published 2026-08-14T16:39:32.000Z | primary source | The Verge | Primary source for the announcement of the watermark removal feature and technical details on SynthID/C2PA. | TechCrunch | Corr
2026-06: OpenAI’s ChatGPT reaches one billion monthly active users 2026-08-11: Google’s Gemini app surpasses one billion monthly active users 2026-08-14: Google announces update to Gemini and Flow allowing removal of visible watermarks
Timeline
OpenAI’s ChatGPT reaches one billion monthly active users
Google’s Gemini app surpasses one billion monthly active users
Google announces update to Gemini and Flow allowing removal of visible watermarks
FAQ
Users can toggle off visible watermarks, such as the 'sparkle' logo, from generated images, videos, and music.
Yes, Google will continue to embed invisible SynthID watermarks and C2PA metadata to verify origin.
The toggle will not launch in countries where visible watermarks are legally required.
Gemini has surpassed one billion monthly active users, placing it on par with OpenAI’s ChatGPT.
Google plans to roll out the setting to Search in the future, but no exact timeline has been specified.
The Nano Banana, Omni, and Lyria models.
Via Settings > Media Watermark in the Gemini app and Google’s Flow video editor.
Yes, invisible SynthID watermarks and C2PA metadata remain active.
An open-source library released by Google for developers to embed local validation mechanisms.
It follows Anthropic’s decision to include watermarks in Claude-generated text to comply with EU regulations.