Why can blacked-out text in a PDF still be read?

Drawing a black box usually adds a shape on top of the text. The text is still stored in the file, so it can be selected, copied, searched or read by an AI tool.

How do I check whether a redaction worked?

Drop the file here and the checker marks any covered text that is still in it. You can also select everything in the PDF, copy it, paste it into a text editor and search for what you meant to remove.

What does the checker look for?

Black boxes, white-out, highlighter and see-through covers over text, redaction marks that were never applied, hidden text layers behind scans, and leftovers from editing in screenshots and photos.

How do I redact a PDF properly?

Use a redaction tool that deletes the text and apply it, or black out the area and save the page as a picture so no text layer remains. Then check the result again.

Is my document uploaded?

No. The file is opened and checked inside your browser tab and is never sent to a server.

Tools › Privacy

PDF & Image Redaction Checker

Blacked something out in a PDF? Check that the text underneath is really gone and can't be copied, searched or pasted into an AI tool. It catches black boxes, white-out, see-through highlighter, hidden text and boxes drawn over scans in PDFs, and leftovers from editing in PNG and JPEG screenshots and photos. You review what it finds and decide what matters.

Animation: a black box, white-out, see-through marker and a box over a scan are each selected, revealing the hidden number underneath

🔒 This tool runs entirely in your browser. Your Document/Image is never uploaded. All logic runs into your browser on your device.

Choose a PDF or Image to check

Drop a PDF or image here, or click to choose PDF, PNG or JPEG: court filings, HR letters, bank statements, screenshots, photos of documents

What this checks, and what to do about a finding

Fake redactions: Black boxes, highlighter or shapes drawn on top of text that is still in the file.
Light or coloured covers: White-out, and grey, yellow or pastel boxes and images over text.
See-through ink: Semi-transparent boxes that dim the text but leave it readable. Normal highlighting is not flagged.
Redaction marks never applied: Redact annotations that were marked but not applied.
Redacted scans: A blacked-out scan whose hidden OCR text layer still has the original words.
Boxes drawn over pictures: A box drawn on top of a scanned page, signature or photo, while the picture underneath is still stored whole in the file. The tool shows what can be extracted from under the box.
Hidden text: White-on-white, microscopic or off-page text, sometimes used to slip instructions to AI screening tools.
Screenshots and photos (PNG, JPEG): Leftover data from an earlier version after the end of the image (the "aCropalypse" bug in some phone and Windows editing tools), colour kept under transparent "erased" areas, boxes that aren't fully solid, an old preview image stored inside a photo, location and device details, and hidden motion-photo video.
How it works: Every page is rendered exactly as a reader sees it, and every character that can be copied out of the file is compared against that picture. Text that can be copied but can't be seen is reported.
Limits of this check: It's automated, so it can miss things (for example text hidden by unusual fonts, words inside pictures, metadata, earlier saved versions or attachments) and it can flag harmless text. It can't read the words inside a scanned picture, so always look through scanned pages yourself. Treat the result as a helper, not a guarantee, and always do your own check before sharing.
If text is still there: Use a tool that actually removes the text (for example Acrobat's Redact, then "Apply"), or flatten the pages to images, then check the new file again.

PDF & Image Redaction Checker

Questions

Related tools