AI Text Cleaner: find and remove hidden characters
Highlights invisible characters, curly quotes, em dashes and odd spaces that machine-written and auto-formatted text carries, and removes them in one click.
Your text
What will be cleaned
Paste text to see what is hiding in it.
Nothing to clean: this text contains no hidden or typographic characters.
What to clean
What this tool is for
Text that came out of a language model or a word processor carries characters you cannot see: zero width spaces, non-breaking spaces, em dashes, curly quotes, a single ellipsis character instead of three dots. This shows every one of them, says what each is and why it stands out, and removes them on request. Be clear about what that does and does not achieve: it makes the text plain and portable, which is genuinely useful. It does not make text undetectable, because detectors judge sentence statistics rather than punctuation.
How to use it
- Paste the text. The highlight panel marks every suspicious character in place, and invisible ones get a visible label.
- Read the table: each character with its code, how many times it occurs and what it will be replaced with.
- Switch off any category you want to keep, for example if the curly quotes are deliberate.
- Copy the cleaned text, or press replace to keep working on it here.
Frequently asked questions
Will this make my text pass an AI detector?
No, and any tool promising that is selling you something. Detectors work on statistical properties of the writing: how predictable each word is, how sentence length varies. Punctuation is not what they measure. What this tool genuinely fixes is text that looks auto-formatted to a human reader, and invisible characters that break code, CSV files and search.
Why do invisible characters matter at all?
They break things quietly. A zero width space inside a password field or a URL makes it fail with no visible reason. A non-breaking space in a CSV column shifts parsing. A soft hyphen appears as a stray hyphen only when the line happens to wrap. Word counts and search stop matching.
Is the em dash really a sign of machine writing?
It is a sign that the text was not typed in a plain editor: models and word processors insert it, most people typing quickly do not. That makes it a hint about the writing process, not evidence about the author. Plenty of careful human writers use em dashes on purpose.
Does anything get sent to a server?
No. The analysis and the cleaning run in your browser, so a draft that is not public yet stays on your machine.