Em dash remover
What the em dash remover does
This em dash remover finds every em dash in your text, the long one at codepoint U+2014, and deletes it. That is the whole operation. It does not swap the dash for a comma, a hyphen or anything else, and there are no settings on this page to change that.
It also touches nothing else. The en dash at U+2013 stays. Hyphens stay. Quotes, ellipsis characters and spacing all stay exactly as you pasted them. A single character in, the same text out minus that character.
Everything happens in your browser. Nothing is uploaded, there is no run limit, no length cap and no account.
Read this first
Why removing the dash can leave two spaces
Most AI text writes the dash with a space on each side: word, space, dash, space, word. Delete the dash and the two spaces are still there, sitting next to each other. "The draft is ready mostly and the edits are small" comes back with a double gap where each dash used to be.
This is the deletion working as specified, not a bug, but it is the thing everyone notices first. There are two straightforward fixes. Run the result through a whitespace tool afterwards to collapse the doubles, or use the em dash replacer instead and set the replacement to a comma or a space, so nothing is left behind to double up.
If your source writes the dash tight against the words, with no surrounding spaces, deletion joins the two words together instead. Worth a read-through either way.
How it works
How to remove em dashes from text in three steps
No options to configure, because the route has none.
- Step 1
- Paste your text into the box. Nothing leaves your browser.
- Step 2
- Press Remove dashes. Every U+2014 disappears, instantly and locally.
- Step 3
- Read the result for doubled spaces and join-ups where a dash used to be, then copy it out.
Picking the right one
Em dash remover or em dash replacer?
Two routes, two different operations. The difference is worth thirty seconds before you paste.
- This remover
- Deletes U+2014 and puts nothing back. No options. En dashes are left alone.
- The em dash replacer
- Swaps the dash for whatever you type in the replacement box, defaulting to a hyphen.
- The replacer also catches en dashes
- It matches both U+2014 and U+2013, so it is the one to use if your text has both.
- Rule of thumb
- Deleting suits tight dashes inside words. Replacing suits spaced dashes, where something has to fill the gap.
The background
Why this character became an AI tell
Two explanations get repeated by people who have looked at it, and both are about the training material rather than anything hidden. Academic writing, journalism and professional publishing use the long dash freely, and that kind of prose is heavily represented in what these models learned from. So the model picked up the mark as a signal of a serious register.
The second is that a model has no mechanism for getting bored of its own punctuation. A human writer who used the same construction four times on a page notices and varies it. A model has no such counter, so a habit that starts reasonable ends up at a rate no editor would allow.
The other side
Is the em dash still an AI giveaway at all?
It is worth knowing that the argument has moved. There is a serious case, made in the tech press this year, that the long dash was never much of a tell and is now useless as one: plenty of human writers have always used it, and the habit is widely enough known that anyone trying to disguise a draft strips them first.
That does not make the tool pointless, it just changes the reason to use it. Removing these dashes is a house style decision. Publications, style guides and editors have preferences, some platforms render the character badly, and plenty of writers simply dislike it. Those are good reasons. "It will stop someone knowing" is not one this page will offer you.
Straight answer
Does removing this one character make text undetectable?
No, and nothing on this page suggests otherwise. Deleting a punctuation mark changes punctuation. It does not change sentence length distribution, vocabulary, paragraph shape or structure, which is where the rest of the pattern lives.
Detection tools report signals, not proof of who wrote something. We are not affiliated with any of them and make no promise about what any of them will say. If a whole document reads uniform, the dashes were the least of it.
In practice
Who uses an em dash remover, and when
Mostly people with a style rule to meet and a deadline in front of them.
- Writers on a house style
- Publications that forbid the character outright, where compliance is a find-and-delete job.
- Developers and documentation writers
- Where a long dash in a code sample, a filename or a terminal block is actively wrong.
- Anyone pasting into a system with poor font support
- Older CMS templates, email clients and legacy exports that render the character as a box.
- Editors doing a fast punctuation pass
- Strip them all, then put back by hand the two or three that were genuinely earning their place.