Gemini Watermarks, SynthID & Content Credentials
How Google documents SynthID, Nano Banana image provenance, Gemini Apps Content Credentials, visible marks, and Google-only verification.
- Published
- Last updated
- Review
- Reviewed against primary sources
Google’s documented provenance stack for Gemini is SynthID-first across modalities, with Gemini Apps also attaching Content Credentials and optionally a visible mark. Those three layers are independent. This hub orients the research cluster. The Gemini watermark pillar answers the generic “Gemini watermark” query in more detail.
Documented provenance systems
- Invisible SynthID in supported generated or edited Google AI media, and in Gemini app/web text.
- C2PA Content Credentials on Gemini Apps media, plus a documented C2PA claim on the Nano Banana 2 Lite API model card.
- A visible Gemini Apps watermark that can be toggled, with regional and account exceptions.
Image provenance
Nano Banana is Google’s Gemini image-generation family, not a single timeless model. Current official names include Nano Banana (Gemini 2.5 Flash Image, legacy), Nano Banana 2 Lite, Nano Banana 2, and Nano Banana Pro. Gemini API docs state all generated images include SynthID. Gemini Apps Help states Apps media includes invisible SynthID plus Content Credentials. Do not copy Apps visible-mark settings onto Studio, the API, Search, or Ads.
Text, video, and audio
DeepMind documents SynthID for Gemini app and web text, for video, and for audio on specified Google surfaces. Public Gemini upload verification covers image, video, and audio for Google AI — not arbitrary pasted text and not other companies’ SynthID. AI Watermark Center does not currently detect SynthID.
Public verification
Signed-in Gemini Apps verification is the current consumer surface. DeepMind’s SynthID Detector portal is still an early-tester waitlist as of this verification date. Google also documents verification through Gemini in Chrome or Search in the Nano Banana 2 Lite announcement.
Official Google verification (opens in a new tab)
AI Watermark Center support
- Embedded C2PA on JPEG/PNG/WebP: available
- SynthID: not available
- Unicode scanner: available, unrelated to Gemini text SynthID
Cluster
Sources
SynthID — Google DeepMind
Accessed August 16, 2026.
SynthID modalities (image, video, audio, text). Gemini Apps can check Google AI media. A separate SynthID Detector portal remains an early-tester waitlist, not a general-availability consumer product, as of this access date.
Verify AI-generated images, videos, and audio — Gemini Apps Help
Accessed August 16, 2026.
Gemini user-facing verification for image, video, and audio. SynthID identification is Google AI only; Content Credentials inspection is a separate layer.
Manage watermark settings in Gemini Apps — Gemini Apps Help
Accessed August 16, 2026.
Current Gemini Apps visible-watermark toggle is Apps-only. India, South Korea, and Vietnam restrict the toggle to AI Ultra; work/school accounts cannot turn it off. The setting does not affect SynthID or Content Credentials.
Nano Banana Pro: Gemini 3 Pro Image model from Google DeepMind — Google
Published November 20, 2025. Accessed August 16, 2026.
Historical 20 Nov 2025 launch post: all Google-tool media embedded with SynthID; Gemini-app image verification. Visible sparkle policy in this post is chronological, not a substitute for current Gemini Apps Help.
Start building with Nano Banana 2 Lite and Gemini Omni Flash — Google
Published June 30, 2026. Accessed August 16, 2026.
Defines the current Nano Banana family (legacy, 2 Lite, 2, Pro) and states Nano Banana 2 Lite and Gemini Omni use SynthID. Mentions verification through the Gemini app, Gemini in Chrome, or Search.
Nano Banana image generation — Google AI for Developers
Accessed August 16, 2026.
Nano Banana is four Gemini API models. All generated images include a SynthID watermark. Does not, by itself, document C2PA for every model.
Gemini 3.1 Flash Lite Image — Google AI for Developers
Accessed August 16, 2026.
Nano Banana 2 Lite (gemini-3.1-flash-lite-image): SynthID always on plus C2PA watermarking. Scope is this API model card, not every Google consumer surface.
Related research
- Gemini watermarkVisible mark, SynthID, and Content Credentials as separate systems.
- Nano Banana watermarkImage-family provenance, not a clone of the Gemini pillar.
- Gemini SynthIDWhere Google products currently use SynthID.
- Check Gemini watermarksGoogle verification versus AWC C2PA inspection.
- Gemini Content CredentialsC2PA on Gemini Apps and documented API models.
- Gemini visible watermarkWhen a visible mark appears, and what it does not prove.
- Nano Banana watermark remover intentInformational. No removal tool.
- Image provenance protocolThis protocol registers how AI Watermark Center will collect owned Nano Banana image samples by Google product surface Study status: dataset pending.
- AI Watermark DatabaseGoogle rows with claim, evidence, and support separated.