AI Watermark Remover & Adder

Remove hidden characters AI leaves in your text

Generated text and some web sources quietly insert zero-width characters — invisible code points like the zero-width space (U+200B), zero-width joiner (U+200D), and the byte-order mark (U+FEFF). They can be used to fingerprint or track copied text, and they break search, diffs, and code. This tool detects every hidden code point, tells you exactly what it found, and gives you a clean copy with one click.

Three modes

• Scan & clean: list and strip invisible characters; unusual spaces are normalized to plain spaces.
• Add watermark: embed an invisible, zero-width signature (any label you choose) into text that looks unchanged.
• Reveal watermark: extract a hidden label back out of watermarked text to trace where a copy came from.

Zero-width & watermark FAQ

Will the watermark survive copy-paste? Usually yes in plain-text contexts, but rich editors that sanitize Unicode may strip it. It is a lightweight signal, not DRM.

Is anything uploaded? No. Detection, cleaning, embedding, and reveal all run entirely in your browser.

Which characters get removed? Zero-width spaces/joiners, BOM, soft hyphens, bidi marks, emoji variation selectors, and odd Unicode spaces normalized to plain spaces or newlines.

Can I fingerprint text I share? Use Add watermark to embed an invisible label, then Reveal watermark on a suspect copy to see if it matches.

Is it free? Yes — no account and no uploads.

Related: AI code detector · Resume de-AI-ifier · AI email rewriter · Base64 decode