How to remove Claude's watermark from Substack
To send a Claude-drafted Substack post without the watermark, rewrite the text through a non-watermarking model before pasting it into the editor, because Substack emails and publishes your exact words. The watermark rides along in the wording to every subscriber and the public archive.
What Substack does with Claude's watermark
Substack preserves pasted text and publishes both a web page and an email. Both are public-facing and permanent, so this is a publish-first surface where cleaning up front is the only practical option.
Steps to clean it
- 1Paste your Claude text into ClaudeRewrite. It rewrites the wording through a non-watermarking model (DeepSeek V4 Flash), breaking the statistical pattern while preserving every fact, number, name and link, then strips any leftover invisible characters.
- 2Paste the cleaned text into the Substack editor with plain-text paste to avoid hidden styling.
- 3Preview the post to confirm formatting, then schedule or send.
- 4Both the emailed and web versions now carry the cleaned text.
FAQ
Can subscribers tell my newsletter was AI-drafted?
Not from the watermark today — there is no public detector. But the archive is permanent and becomes checkable when detection ships, so clean before you send.
Remember: Claude's watermark is statistical — it lives in the word choices, so it travels with your text into Substack no matter how you paste. There is no public detector today, but published text stays checkable once Anthropic's detection API ships. See which models are marked →