Claude · Anthropic · SynthID-Text

Claude’s watermark is not hidden ink. It is which word came next.

Anthropic published the method. Future Claude models pick each next word with a secret key and the words so far. The sentence looks ordinary because it is. A key holder can ask whether Claude was likely involved. Nothing is added to the text.

Reviewed by · Published

or type below, or drop a file

Inspected in your browser. A plan is required to finish.

No text? Try one of these:

Choose the surface, then the tool

SurfaceGifi actionWhat can be verified
Ordinary wordsRewrite the proseThe words changed; no private detector certificate
C2PA, EXIF, XMP, propertiesStrip named metadataAction list plus an independent reinspection
Pixels, audio, or videoNot supportedNo removal claim

The dice are keyed

The method is SynthID-Text, from Google DeepMind. When two words would do, the key settles which one. Light edits usually keep the trail. A full rewrite does not. Short text, code and tight facts have fewer forks, so the mark is weaker there.

A hit means Claude was likely involved. It does not mean a person never touched the page. Anthropic will ship a detector. Older models are still in transition.

The file is only a note

Supported images and documents get a C2PA credential — a signed note in metadata, not a pattern in the words. We can strip the local label. Soft binding can survive.

What this cannot promise

  • A rewrite does not prove the text was never AI-assisted.
  • No one outside the lab can certify their detector. The key is private.
  • Pixel watermarks, audio and video are out of scope.

Questions this usually raises

Will a light edit remove it?
Usually not. Anthropic said so. Replacing the wording does.
Can it identify me?
No. It tests whether Claude was involved, not who asked.
Can you certify Anthropic’s detector?
No. The key is private.

Claude

Statistical signal? Rewrite the prose.

A Claude mark, if it is in the words, needs different words. Hidden characters and C2PA need the file cleaner.