Simple Unmark guide
Does Claude Watermark Text?
- Published
- Last updated
- Sources checked
- Primary references: 3
Short answer
Yes. Anthropic documents that supported Claude models weave a machine-readable watermark into generated text, applied globally. It is a keyed token-choice method, a version of the SynthID-Text approach, with nothing added to the text and no hidden characters. Light editing probably will not remove it; a complete rewrite will.
This changed in August 2026. Before then, no first-party Anthropic documentation established text watermarking. Anthropic has now published how it works, so the answer is a documented yes.
Choose the privacy boundary before every rewrite.
Private is the lowest-cost path and stores no text in our application database. Confidential verifies Google Confidential Space in your browser and encrypts to a key the workload created for that one request, so our web backend receives no text — though DeepInfra still performs the rewrite. Confidential AI, with the model inside a confidential GPU, is coming soon.
Compare the three modes and their limitsWhat Anthropic documents
Anthropic describes weaving an imperceptible watermark directly into generated text, without changing the meaning, quality or readability of the response. Instead of arbitrary randomness, the key plus the preceding words settle which word the model picks — and Anthropic identifies this as a version of the SynthID-Text approach published by Google DeepMind in Nature in 2024.
The scope is broad. Anthropic states it is applying watermarking globally at launch rather than scoping it to the EU, in the context of transparency obligations under the EU AI Act. Models launched before 2 August 2026 were to be updated over the following months. Separately, supported file types can carry cryptographically signed C2PA provenance metadata.
What Anthropic says removes or weakens it
Anthropic's own description is the clearest statement available of what actually affects this class of watermark.
- A complete rewrite in which every word is replaced will remove it; light editing probably will not.
- Text Claude only proofread carries very little, because nearly all the words are the person's.
- Factual passages carry a sparser mark, because there are fewer choices to encode into.
- Code generally carries less, since it often has to be exact.
- Detection does not work well on small samples.
- Translations do carry the watermark, because every word is chosen by Claude.
What it cannot establish
Anthropic states that a watermark can only determine that Claude was likely involved with the content at some point, and cannot distinguish "Claude wrote this" from "Claude heavily edited this". It says nothing about ownership or authorship and does not change a user's rights.
Anyone treating a watermark hit as proof that a person did not write something is asserting more than the mechanism supports — and, given that proofreading leaves almost no mark while translation leaves a full one, the mapping between the signal and human effort is genuinely loose.
A watermark can only determine that Claude was likely involved with the content at some point. It cannot distinguish 'Claude wrote this' from 'Claude heavily edited this.'
Can you check it yourself?
Anthropic offers a watermark detection API in restricted private preview to eligible organizations and enterprises with EU compliance obligations. Its documentation notes an API update dated 1 September 2026. This is not a general public checker, and Simple Unmark does not have access, so we cannot measure the remaining watermark signal after rewriting.
Sources
- How Claude's text watermarking works
Anthropic · First-party source
- Transparency Hub
Anthropic · First-party source
- Scalable watermarking for identifying large language model outputs
Nature (2024)
Published by Simple Unmark. Our Methodology explains how we source claims, what we test, and what we do not claim.
Apply this to your own text.
Guests get three free cleans without an account. Choose Private or browser-verified Confidential processing before each rewrite. Deterministic character inspection is free and runs only in your browser.
Keep reading
- How AI Text Watermarks WorkHow hidden-character marks and statistical token watermarks differ, how detection works, and why removal claims should always be bounded.
- What Is SynthID Text?SynthID Text explained: how Google's text watermark is applied during generation, how detection works, what weakens it, and how it differs from image watermarking.
- Hidden Unicode vs Statistical Text WatermarksThe two mechanisms confused most often: invisible Unicode characters versus keyed token-choice watermarks. What each one is, what removes it, and what neither proves.
- AI Watermarks vs AI Detectors vs Provenance MetadataThree different things routinely confused: keyed watermarks added at generation, statistical AI detectors that guess, and C2PA metadata attached to files.
