← All posts

Is it legal to remove a PDF password?

You have a PDF in front of you that will not do something. It will not print, or it will not let you copy a paragraph out, or it asks for a password every time you open it. There is a website that says it will take that off in ten seconds.

Before you paste your file into it, you would quite like to know whether you are about to do something you should not. Not in a hand-wringing way. You just do not want the answer to turn out to be "that was a criminal offence" three months later.

Nobody has written that answer. The top result for this question is a forum thread from 2011, and the commercial pages below it dispose of legality in one line on the way to the download button. So here is the shape of it, from someone who is not a lawyer and is not giving you legal advice: what the two different acts actually are, why the difference matters more than it looks, and which layer is the one that usually catches people out. It is not the one you are worried about.

There are two acts here, and only one of them is getting in

A PDF can hold two kinds of password, and almost every argument about this question is two people describing different files.

The user password is the one that makes the document readable at all: everything in the file is ciphertext until it is supplied, which is why the prompt keeps coming. Removing it when you have it is decryption you were already doing every time you opened the document. Getting past it without the password is not removal, it is guessing.

The owner password does not close anything. This is the part worth knowing, because it changes what "removing the protection" even means. When a file has only an owner password, it is encrypted with a key derived from the empty user password. Every conforming reader on earth already computes that key and decrypts the document without asking anybody for anything. The restrictions — no printing, no copying, no editing — are a handful of flags in the encryption dictionary, and they are honoured because readers volunteer to honour them.

So when an owner restriction comes off, nothing has been broken into. A program has simply stopped volunteering. The content was never withheld from you at any point in the process; it was decrypted on your screen the whole time.

Which one you have takes a second to tell: a prompt before you can see anything means the first kind, and a file that opens instantly and then greys out Print means the second. Our PDF inspector will confirm it from the file itself, in the browser, with nothing uploaded. For the mechanics of either case, and for the procedure, see why a PDF asks for its password every time and removing a password you already know; this post is about whether you should.

Why the distinction is not something we invented

It maps, almost uncomfortably neatly, onto how the law is written.

In the United States, the DMCA's section 1201 has two halves. Subsection (a) prohibits circumventing a technological measure that effectively controls access to a copyrighted work — and the act itself is prohibited, not only the tools. Subsection (b) deals with measures that protect a right of the copyright owner, such as copying, and it prohibits trafficking in the tools. There is no matching provision making the act of circumventing a use control unlawful in itself. Congress left that gap deliberately, because the person doing it might be exercising fair use.

A user password is an access control. An owner restriction is a use control. The statute's own division and the PDF specification's own division are the same division.

In the UK, section 296ZA of the Copyright, Designs and Patents Act gives the person who applied an effective technological measure the same civil remedies as against an infringer. It is a civil claim by that person, not a crime, and the criminal provision next to it — 296ZB — is aimed at making, selling or advertising the means, not at the individual act. The EU's InfoSoc Directive works similarly, and both turn on the word "effective". Do not build anything on the argument that a permission flag any reader can ignore is not an effective measure. It is a tempting argument, courts have not been generous with it, and you do not want to be the test case.

There is also a route most people never hear about. UK law contains permitted acts that a technical measure can block — making an accessible copy for a disabled person is the clearest — and section 296ZE exists precisely so that the measure is not allowed to defeat the permitted act. In the US, the Copyright Office's recurring exemptions have long included works needed for use with assistive technology. The PDF format agrees with them: it carried a separate permission bit that meant "extract text for accessibility" as distinct from ordinary copying, and PDF 2.0 deprecated that bit entirely, so conforming readers now treat accessibility extraction as always permitted. When a screen reader cannot read a document because copying is switched off, nobody's law is on the side of the flag.

For most people who type this question, no statute is ever going to come near it. What comes near it is a contract.

An employment policy about copies of client files. An NDA on a data room whose documents are all issued with restrictions. The licence on a paid market report that says one printed copy. A court's rules on filed documents. The terms you accepted with a portal.

None of these care which password you had. They care what you did with the contents and whether you were allowed to hold a copy in that form. And they bite far earlier and far more cheaply than copyright law does — a disciplinary process, a terminated licence, an unenforceable submission. Reading the policy takes ten minutes and answers the question you were actually asking.

The other thing worth ten minutes: asking. A great many restricted PDFs are restricted because a document management system stamps every export the same way, and nobody chose it for your file. The person who sent it can usually send it again without.

What the tool checks, and what it cannot

Our Unlock PDF tool checks exactly one thing: that the password you typed decrypts the file. If it does not, the job stops and says so. There is no guessing, no dictionary and no attack on the encryption, so a document whose password you do not have stays closed no matter how many times you try it.

That check is a technical fact, not permission. The tool has no idea who you are, whose document it is, or what you agreed to when you were given it. It cannot make a judgement about entitlement and it does not pretend to. Our terms require you to have the right to the files you upload, and that requirement is doing real work — it is the part the software cannot do for you.

What it does do is honest about the result. The file comes back with no security dictionary at all, the notes list which restrictions were lifted so you can see what was on it rather than guessing, and the name gets -unlocked appended so you do not confuse the two copies in a folder. On an owner-restricted file you leave the password box empty, because there is nothing to type.

Doing it locally, which is free and takes one line

qpdf is open source, installs from any package manager, and never touches the network:

pdftk locked.pdf input_pw SECRET output open.pdf is the equivalent for anyone who already has pdftk installed, and macOS Preview will export an unencrypted copy of a file you can open, with no install at all.

If the honest answer is "more secure, not less"

Sometimes the reason you were reading about removing protection is that you are on the other side of it, and what you want is for a document to stop being altered.

Password protect PDF encrypts with AES and sets the flags — but be clear-eyed about what you are buying. The restrictions are the same request a reader chooses to honour, described above, so setting "no editing" on a file and sending it to somebody determined achieves very little; only the opening password withholds anything. There is a trap in the passwords themselves, too, and putting a password back on sets it out: leave the permissions password blank and your recipients inherit the power to lift whatever you set.

If what you actually want is for filled-in answers to stop being re-typed, encryption is the wrong instrument entirely. Flatten PDF stamps the values into the page as ordinary content and deletes the field objects, so there is nothing left to edit in any reader, with or without a password. It cannot be undone in that file, so keep the original.

And if you only landed here because a tool refused your file, the same two-passwords distinction is what decides it — whether a protected PDF can be compressed turns on exactly the same fork in the road.

The short version

Taking the password off your own bank statement, your payslip or a report you were sent to work with is unremarkable, and nobody is coming. Lifting an owner restriction so a screen reader can read a document is a case the law has explicitly thought about and mostly sided with you on. Getting into a file you were never given the password to is not a grey area, and no tool that is honest about what it does will help you with it.

Everything in between is answered by the document and your relationship to it, not by the software. The tool being willing to proceed is not permission. It never was.

Tools this post uses: