Why black boxes don't hide text in a PDF, and how to redact properly

A black box drawn over text in a PDF hides the text on screen only. The words stay in the file, underneath the box, and anyone can select them, copy them or search for them. To redact properly, remove the text itself, then check the output by searching and copying the area you hid. Use the redact tool, which removes the text on the pages you mark, instead of drawing shapes over it.

Redact PDFHide private text in a PDFDo it free

Why the text is still there

A PDF page is a set of drawing instructions. Text is stored as characters with positions on the page, and the black rectangle is another instruction drawn after the text. The viewer then paints the rectangle over the words. The words are still in the file, and a viewer that ignores the rectangle can show them.

Drawing a black rectangle in Word, Google Docs or a PDF editor has the same effect. Saving or printing to PDF from a word processor often keeps the hidden text as well, because the text was never removed from the document.

Real cases where it failed

These are cases that were widely reported.

The Manafort filing, January 2019. Lawyers for Paul Manafort filed a court document with black bars over passages they wanted kept private. The bars sat over the text, and the words under them could be copied and pasted into another document. Several news outlets reported the mistake in January 2019. Sources differ on who spotted it first, so the reports do not agree on that detail.

The Calipari report, 2005. A U.S. Army report on the death of Nicola Calipari, an Italian intelligence officer killed in Baghdad, was published with the classified passages blacked out. An Italian blogger copied and pasted the hidden text. A news report published on 1 November 2005 said the Army investigation concluded that human error was to blame. Army Lt. Col. Steven A. Boylan said the people involved had "misunderstood the capability of the Adobe program." The lesson is the same as in 2019: the black bars were drawings, and the text under them was intact.

These cases are a reminder that the person who sees the black box is not always the person who removed the text. Once a file is shared, anyone can copy the words beneath a box.

Redact the text, not the picture

Open the redact tool and choose one of the two ways to mark the content:

1. Draw a box over the passage you want removed.
2. Search for a word or number, such as a name or an account number, and mark every match.

When you save, the tool treats each page with a mark as a picture and removes the text on it. The pages with marks are saved as images, so the hidden words are no longer in the file. Pages with no marks keep their text, so they still search and copy normally. The tool also strips the document's metadata, which can hold the author's name, edit history and earlier versions of the file.

The trade-off is that a marked page becomes a picture. You cannot search or copy its text afterward, and the file may be a little larger. For a document that is going to be shared, that is the point.

Check the redacted file

Do not trust the look of the page. Check the saved file with four tests:

1. Search. Open the new file, press Ctrl+F (Cmd+F on a Mac), and search for a word you removed. If it finds it, the text is still there.
2. Select. Try to select and copy the text across the marked area. Nothing should come out.
3. Extract. Run the PDF to text tool on the redacted file and search the output for the removed words. The text output shows everything the file contains that a reader could pick up.
4. Reopen. Open the file in a second viewer and check the marks are still covering the right text, and that no page has been missed.

Keep the unredacted original somewhere safe and never send it. Save the redacted copy under a new name, so you cannot mix the two up later.

Images and screenshots

The same rule applies to pictures. A black box drawn over a screenshot leaves the picture underneath, and a blurred or pixelated area can still show a shape or a pattern, depending on how it was made. A solid fill removed from the image itself is the safer choice. The redact image tool does that. The screenshot-specific steps are in the guide to redacting screenshots.

If the screenshot contains text, copy the text from the picture only to check what you are hiding. Then redact the picture itself.

Mistakes that leave text behind

• Drawing a shape over the text in any editor instead of redacting it
• Redacting the visible page but leaving form fields, comments or notes filled in
• Saving a copy with a different name but leaving the original unchanged in the same folder
• Relying on a single search: check the output by searching, copying and extracting the text

Quick answers

Is a black highlight the same as redaction? No. A highlight is a drawn colour over the text, and the text remains.

Does printing the file to PDF remove the text? Not by itself. Printing to PDF usually keeps the text under any shape, so check the output with the tests above.

What if I can still see the text in the file? Redact the page again with the redact tool, save the output, and run the checks again. Do not send the old copy.