Zero width space remover
What the Perplexity zero width space remover does
Paste an answer copied out of Perplexity and this tool strips the invisible Unicode characters it is carrying. Fourteen of them, named below with their codepoints. The zero-width family and the directional controls are deleted outright. The three space look-alikes become ordinary spaces, because deleting them would join two words together.
Everything else in your text is left exactly as it is. Line breaks stay, blank lines stay, run spacing stays, Markdown stays, and your quotes and dashes are not rewritten. This route does one job, and the reason to use it is that you know what it will and will not touch.
Read this first
Deleting a Space in Perplexity is a different thing
Search for this and half the results are about the other meaning. Perplexity has a feature called Spaces: named workspaces that hold threads, files and instructions. Deleting one of those is something you do inside Perplexity's own settings, and no tool on the open web can do it for you.
This page is about the other kind of space, the character. Zero-width spaces, hair spaces and no-break spaces that ride along in the clipboard when you copy an answer. If you came here to remove a workspace, this is not the page.
The characters
The invisible characters this remover strips
Fourteen characters, and the list is the whole list. Two different behaviours, so they are grouped by what happens to them.
- Deleted: U+200B, U+200C, U+200D, U+2060
- Zero-width space, non-joiner, joiner and word joiner. Removed completely. Joiners that build an emoji are recognised and kept.
- Deleted: U+FEFF
- Byte-order mark. Fine at the very top of a file, a problem anywhere else in one.
- Deleted: U+00AD
- Soft hyphen. Shows nothing until a line wraps, then puts a hyphen in a word that never had one.
- Deleted: U+202A to U+202E
- Five directional controls. Embeddings, the pop, and the two overrides. Stray ones reorder mixed-direction text.
- Replaced with a space: U+00A0, U+202F, U+200A
- No-break space, narrow no-break space and hair space. Each becomes a normal space, so words stay separated.
How it works
How to remove zero width spaces from Perplexity text in three steps
No account, no upload, nothing to install. The whole thing runs in the page you are reading.
- Step 1. Copy the answer
- Take it straight from Perplexity, citation markers and all. Do not paste through a plain-text editor first or you will strip the characters before the tool can show you them.
- Step 2. Paste it in
- Drop the text into the box. The count of characters removed or changed appears as soon as it runs.
- Step 3. Copy the clean version
- Take the output and paste it into your document, your CMS or your editor. The wording is identical to what you put in.
Where they come from
Why Perplexity answers carry zero width spaces
Perplexity does not write an answer from nothing. It fetches web pages, quotes them and summarises them. Those pages were typeset by people and by publishing systems, and both put invisible characters in on purpose: a narrow no-break space before a unit so it cannot wrap, a soft hyphen inside a long term so it breaks tidily, a byte-order mark added by whatever exported the block.
Once that material is carried into an answer and rendered as styled text in your browser, the invisibles go into the clipboard with the words. A model answering from memory with no retrieval step has one source of residue. A retrieval engine has two.
Perplexity also lets you choose which model writes the answer, so the same account can produce answers that behave differently. Inconsistent results between sessions are normal and not a fault in the scan.
Why it matters
What a zero width space actually breaks
It is almost always found as a bug rather than as a privacy concern. A zero-width space inside a heading splits the keyword, so the slug your CMS generates stops matching the title a reader sees. A no-break space inside a shell command or a spreadsheet formula produces an error that points nowhere useful.
The same characters make find-and-replace in Word miss an obvious match, make a CSV import reject a column that looks correct, make a screen reader pause in the middle of a word, and make a git diff show a changed line where nothing visible changed. None of it looks like a Unicode problem until you check.
Straight answer
Does removing zero width spaces hide that I used AI?
No, and it is worth being blunt about it. AI detectors score sentence structure, word choice and rhythm. None of those changes when an invisible character is deleted. If a detector flagged your text before the clean, the same input will read the same way afterwards.
The honest reason to run this is mechanical. Invisible characters break things downstream, and text that goes into a CMS, a repository or a data file is better off without them. That is a real problem with a real fix. Beating a detector is not what this tool is for, and no page should tell you otherwise.
In practice
Who uses the Perplexity zero width space remover
Writers moving research out of Perplexity and into a CMS, where a split keyword in a heading shows up later as a mismatched URL. Analysts pasting figures into spreadsheets, where a no-break space turns a number into text. Developers dropping a config snippet from an answer into a file that then refuses to parse.
It is also the first thing to try when a document misbehaves for no visible reason. If the text looks right and the tool reading it disagrees, an invisible character is a good early suspect.