Will a light edit remove it?

Usually not. Anthropic says light editing probably keeps the signal. Replacing every word does not — though at that point the text is no longer the original generation.

What about short text?

Fewer forks mean less evidence. Detection is weak on short samples and grows more confident as the passage lengthens.

Code has fewer equally-good next tokens, so the mark is sparser. Comments and other arbitrary wording can still carry it; a required identifier cannot.

What if Claude only proofread?

The watermark only attaches to words Claude chose. Light grammar edits may be too few to register. Heavier rewriting leaves more of a trail.

Do translations carry it?

Yes. A translation produced by Claude is a sequence of words Claude chose, so the watermark has a full passage to live in.

No. Nothing in the watermark or its key points to a person, organisation, or chat. It tests whether Claude was involved, not who asked.

What about other models?

Other providers who signed the same EU Code of Practice will use their own keys, and possibly their own methods. A Claude check says nothing about them.

What does a hit prove?

That Claude was likely involved at some point. It cannot tell “Claude wrote this” from “Claude heavily edited this,” and it does not prove human authorship.

What about images and files?

Those get a C2PA content credential in metadata — a signed note, not SynthID-Text. The file cleaner can strip that label. Soft binding can survive and re-link to a remote manifest.

Claude · SynthID-Text · EU AI Act

Markdown

Claude’s watermark is not hidden ink. It is a trail of choices.

Anthropic will watermark future Claude models so a key holder can test whether Claude was involved — a requirement of the EU AI Act. The method is SynthID-Text. Nothing is added to the text.

The demonstrations below illustrate that mechanism. They do not reproduce Anthropic’s key or detector.

One prompt can take many valid paths

Claude writes one token at a time. When several next words are equally good, a random number chooses among them. An early choice changes the context for every choice that follows.

promptWrite one moody opening line.

Generated path

The rain before anyone thought to run.

At this fork, the model’s shortlist was:

42%

arrived without warning,

34%

started before dawn,

24%

came in from the sea,

The watermark changes the dice, not the vocabulary

SynthID-Text, the method Anthropic uses, derives each random number from a secret key and the words written so far. The shortlist can stay the same while the sequence of choices leaves a pattern the key holder can test.

Ordinary generation

numbers from: fresh random numbers
numbers used: 0.18 · 0.71 · 0.43

Keyed generation

numbers from: key + words so far
numbers used: 0.18 · 0.71 · 0.43

To disturb that sequence, choose a new path

Rewrite depth

More new phrasing usually means more new token choices — and more risk of changing tone.

Watermarked path

The rain arrived without warning, turning the avenue silver before anyone thought to run.

New path

Without warning, rain glazed the road in a sheet of silver and caught the pedestrians still walking.

Wording changed57%

Illustrative keyed match31%

Illustrative only. A real score depends on Anthropic’s key, detector, text length and implementation. Gifi cannot see or certify any of those.

Deleting characters does not touch Claude's text mark

Claude's text watermark adds nothing you can select and delete. C2PA is what Claude attaches to images and documents — a signed note in metadata, a different job. A statistical mark is a property of the choices across the text.

Deterministic

Hidden characters and C2PA

U+200B · U+E0067 · EXIF · C2PA

We can name the exact carrier, remove it, and show a precise before-and-after report. The visible sentence can stay unchanged. This is not Claude's text watermark.

CleanFree in your browser

Probabilistic

Claude's token-choice pattern

SynthID-Text · key + words so far

There is no stray character to delete. Rewriting samples a new sequence, but every edit carries some risk of meaning or tone drift.

RewriteModel call · paid

Detection needs room for chance

More high-entropy choices give a detector more evidence.

illustrative capacity

Open prose

strong

Summary

moderate

Code

weak

Exact quote

none

Very short text, exact quotations and constrained outputs may not contain enough choices for reliable attribution in the first place. A watermark check is evidence, not an identity test.

From Anthropic’s own questions

What the watermark does, and does not, mean

Will a light edit remove it?: Usually not. Anthropic says light editing probably keeps the signal. Replacing every word does not — though at that point the text is no longer the original generation.
What about short text?: Fewer forks mean less evidence. Detection is weak on short samples and grows more confident as the passage lengthens.
What about code?: Code has fewer equally-good next tokens, so the mark is sparser. Comments and other arbitrary wording can still carry it; a required identifier cannot.
What if Claude only proofread?: The watermark only attaches to words Claude chose. Light grammar edits may be too few to register. Heavier rewriting leaves more of a trail.
Do translations carry it?: Yes. A translation produced by Claude is a sequence of words Claude chose, so the watermark has a full passage to live in.
Can it identify me?: No. Nothing in the watermark or its key points to a person, organisation, or chat. It tests whether Claude was involved, not who asked.
What about other models?: Other providers who signed the same EU Code of Practice will use their own keys, and possibly their own methods. A Claude check says nothing about them.
What does a hit prove?: That Claude was likely involved at some point. It cannot tell “Claude wrote this” from “Claude heavily edited this,” and it does not prove human authorship.
What about images and files?: Those get a C2PA content credential in metadata — a signed note, not SynthID-Text. The file cleaner can strip that label. Soft binding can survive and re-link to a remote manifest.

Choose the right tool

Statistical signal? Rewrite the prose. Hidden characters or C2PA? Clean the file.

Gifi keeps those jobs separate because they have different costs, different risks, and different standards of proof.

Rewrite the words How file marks work Clean hidden marks

Use rewriting only on content you own or are authorised to process. A changed statistical signal does not prove human authorship, and no third-party tool can guarantee the result of a private detector. Read Anthropic’s explanation.