GPT Watermarker

SynthID Detector

Check AI images for Google's invisible SynthID watermark and the C2PA manifest. C2PA is read here in your browser; SynthID is confirmed with the official detector — the only place it can be. Gemini, Nano Banana & ChatGPT.

SynthIDC2PA

Detect hidden watermark

Check for C2PA in your browser, then verify Google's SynthID with the official detector.

SynthID Detector: Check AI Images for Google's Watermark

SynthID is Google DeepMind's invisible watermark, woven into the pixels of AI images from Gemini, Nano Banana, Imagen — and present alongside C2PA on ChatGPT images too. It's designed so only the issuer's own detector can read it back, which is why no honest third-party tool can "scan for SynthID" on its own. This SynthID detector does the part it can prove — reading the file for a C2PA Content Credentials manifest instantly in your browser — and then hands you to the official provenance checker to confirm SynthID itself, step by step. If the image is watermarked, you proceed straight to the SynthID remover to clean both the pixel signal and the C2PA layer.

Why a 'SynthID Scanner' Can't Be Instant

SynthID embeds a statistical pattern across an image's pixels that its detector model is trained to recognize. By design, that pattern can't be read by anyone without the model — so any browser widget claiming to instantly "detect SynthID" is really just checking metadata and relabeling it. We don't do that. This tool is explicit about the boundary: C2PA we read ourselves; SynthID we verify through the official checker, because that's the only place a real answer comes from.

The Honest Two-Step Check

First, the detector reads your file for a C2PA manifest. If it's there, the image is provably AI-generated and you can clean it immediately. If C2PA is clear, that's not the end — a Gemini or ChatGPT image can still carry SynthID in the pixels. The tool opens the official content-provenance checker, you upload the same image there, and if it reports an AI watermark you come back and proceed to clean. Two layers, each checked where it can actually be confirmed.

Removing SynthID Once Confirmed

Confirmed SynthID needs frequency-domain cleaning, not a metadata wipe — the signal lives in the pixels and survives crops, screenshots and re-saving. The remover this detector routes to disrupts that signal where it lives and strips the C2PA manifest in the same pass, returning a clean file with no visible change. Which AI images carry SynthID: Gemini (including Nano Banana / Nano Banana Pro), Imagen, and ChatGPT stills.

SynthID Detection FAQs

Straight answers on what each workflow removes, how files are handled, and what result you should expect.

Can you detect a SynthID watermark?

Only Google's own detector can read SynthID back — that's how the watermark is designed. This tool confirms C2PA itself and hands you to the official provenance checker for SynthID, then lets you proceed to clean if it's present. Any tool claiming instant SynthID detection is checking metadata, not SynthID.

Which AI images use SynthID?

Google's image stack — Gemini, the Nano Banana and Nano Banana Pro models, and Imagen — and ChatGPT stills carry a SynthID signal alongside C2PA.

If there's no C2PA, is the image clean?

Not necessarily. SynthID lives in the pixels and survives a metadata strip, so a clean C2PA result on a Gemini or ChatGPT image still warrants checking SynthID with the official detector.

How do you remove SynthID once it's confirmed?

With frequency-domain cleaning that disrupts the pixel-level signal, plus a C2PA strip in the same pass. A plain metadata wipe does not remove SynthID.

Is the detection private?

The C2PA read happens in your browser and doesn't upload your file. The SynthID step uses the official checker, which you upload to separately and by choice.
    SynthID Detector — Check AI Images for Google's Watermark | GPT Watermarker