Provenance Lens

How to Detect a SynthID Watermark

Learn how Google currently checks SynthID in supported media, why SynthID differs from C2PA and what Provenance Lens can and cannot verify today.

Updated: Published by Provenance Lens

Use Google's documented verification path

Google currently directs people to Gemini to check supported images, videos and audio for SynthID. Its separate SynthID Detector portal remains an early-tester programme for journalists and media professionals.

Provenance Lens status: UNSUPPORTED

Provenance Lens does not implement Google's proprietary SynthID detection mechanism. It reports that check as UNSUPPORTED. It does not substitute a visual classifier or turn an unavailable check into NOT_DETECTED.

SynthID is not C2PA

SynthID embeds a watermark into generated media. C2PA carries signed metadata with richer provenance context. One may remain when the other is absent, so the two signals need separate checks and separate interpretation.

What a report cannot prove

  • A valid signature proves who signed a manifest, not that the manifest's claims are true.
  • Metadata is trivially editable. Treat generator fields as an indication, never as proof.
  • Hidden Unicode characters have many innocent causes, including ordinary copy-paste.

Primary sources