Free · no sign-up · your file stays in your browser
SynthID checker
SynthID is an invisible watermark that Google and OpenAI hide inside AI images and audio. Drop a file to check for OpenAI's mark, and see its Content Credentials too.
What we check, in order
Step 1
Signed credentials
If the file was made by ChatGPT, Gemini, Adobe Firefly and others, it often carries signed Content Credentials (C2PA). We read them on your device. A valid signature is very hard to fake.
Step 2
File metadata
EXIF, XMP and IPTC fields can name the camera or the AI tool. Stable Diffusion and ComfyUI often save the prompt inside PNGs. Useful, but anyone can edit these.
Step 3
Invisible watermark
If there are no credentials, we ask OpenAI's official checker to look for its invisible SynthID watermark, which survives screenshots and re-saving.
Step 4
Honest when it's unclear
When none of this turns up anything, we say so. No signal is not proof of a real photo. We'll never show a made-up “97% AI” score.
How to read the result
- Made with AIA signed credential or a watermark says AI. Strongest evidence there is.
- Signs of AIUnsigned metadata names an AI tool. Usually right, but editable.
- Signed camera photoSigned camera credentials and no AI signals.
- Not enough evidenceNothing we can read either way. Common for screenshots and reposts.
Who can read which watermark
| Made with | Watermark | Checked here? |
|---|---|---|
| ChatGPT, GPT Image, OpenAI audio | SynthID (OpenAI key) + C2PA | Yes, via OpenAI's official checker |
| Gemini, Nano Banana, Imagen, Veo | SynthID (Google key) + C2PA | C2PA yes. SynthID: use Gemini's own check |
| Adobe Firefly | C2PA | Yes, on your device |
| Midjourney, Grok, Stable Diffusion | None, or plain metadata | Metadata only, if it's still there |
Questions
- What is SynthID?
- SynthID is an invisible watermark made by Google DeepMind. It's mixed into the pixels (or the sound) when an AI tool creates the file, so it can survive screenshots, crops and re-saving that wipe normal metadata.
- Who uses SynthID?
- Google's tools (Gemini, Nano Banana, Imagen, Veo, Lyria) and, since 2026, OpenAI's ChatGPT images and audio. Each company can only read its own SynthID marks.
- Can you check Google's SynthID too?
- We check OpenAI's SynthID through OpenAI's official checker. Google doesn't offer a public way for other sites to check its SynthID yet, so for Gemini images we point you to Gemini's own check. We'd rather say that than pretend.
- What does “no OpenAI watermark” mean?
- It means OpenAI's checker found no OpenAI watermark. The image could still be from Midjourney, Gemini, Stable Diffusion or another tool, or it could be a real photo.
- Can you remove SynthID?
- No. We don't help remove watermarks. They exist so people can tell what's real.