Does Claude watermark text? (September 2026)
By Claude Watermark Research · Updated
Yes. Anthropic embeds an imperceptible statistical watermark in text from Claude models launched on or after 2 August 2026. Claude Fable 5.1 and Claude Mythos 5.1 launched on 1 September 2026 and are explicitly listed as marked. Opus 5, Sonnet 5, Fable 5, Mythos 5, Opus 4.8 and Haiku 4.5 launched before the cutoff and remain unmarked while Anthropic works through its transition period. The status is model-specific and can change, so this page is dated. Checked against Anthropic's help centre on 2026-09-02.
Check and remove AI traces from my text. Checking runs in your browser, no account, nothing uploaded.
What the mark is
It is distributional, not metadata and not hidden characters. During generation the model's choice among near-equivalent next tokens is biased by a secret key, producing a statistical signature detectable across a long enough passage. The technique is the same family as Google's SynthID-Text tournament sampling.
Because the signature is carried by word choice itself, it survives copy-paste, and no character-stripping tool can touch it. Tools advertising “Claude watermark removal” by deleting zero-width characters are addressing a different thing entirely.
Scope and limits
The policy is global, driven by Article 50 of the EU AI Act, and applies across claude.ai, the API, Claude Code, Cowork and Tag. There is no opt-out and no paid tier that disables it.
Anthropic states plainly that a detected mark shows content “may have been processed by Claude” and is “not fully conclusive”, and that a missing mark proves nothing. It also notes marks may not survive heavy editing, paraphrasing, translation, format conversion or being mixed into other writing.
Detection is in private preview
Anthropic's detection API is available in private preview to eligible organisations, including researchers, educational organisations and enterprises with compliance obligations. There is no public detector, so public tools cannot verify whether a passage carries the mark or certify that a removal worked.
Questions
- Is Claude Fable 5.1 text watermarked?
- Yes. Anthropic explicitly lists Claude Fable 5.1 as a supported marked model, and marking applies wherever the model is used.
- Will paraphrasing remove it?
- Anthropic says light editing probably will not remove the mark completely, while a complete rewrite in which every word is replaced will. Public tools cannot confirm the result while detector access remains private.
Check your own text. Free, unlimited, no account, and it runs in your browser so nothing is uploaded.
Check and remove AI traces
