Why black boxes on a PDF leak
Adding a black rectangle to a PDF may change its appearance without removing underlying text. Remaining text can sometimes be copied, searched or extracted by a parser. OCR reads visible image text; it does not establish that the content beneath a dark box was removed.
Real-world failures
Paul Manafort's lawyers covered sensitive passages in a court filing with black boxes — reporters copy-pasted the "hidden" text right out.
Such failures show why the visible appearance of a black box is not enough to establish that sensitive content was removed from the output.
Why it happens
"Draw a rectangle" in graphics software just adds an opaque shape on top of the page; it deletes nothing underneath. The text, images, and vector paths all remain in the file.
Check what the tool changes and what remains in the downloaded output, rather than relying on a button label.
What real redaction is
PrivScrub rebuilds every output page as an image in a new PDF, with selected regions covered in its pixels. No original pages or resources are copied. It rereads the output for the listed checks, but does not run OCR.
Unboxed areas retain their visible appearance, but every page loses searchable and accessible text and needs visual review.
Don't forget to strip document metadata (author, timestamps) too, or identity info still leaks.
Current PrivScrub PDF scope
PDF editing supports basic, unencrypted files up to 50 pages and 25 MiB. Nonempty annotations, internal or external links, forms, attachments, actions and layers are unsupported and stop processing before export. Redaction rebuilds every page as an image, removing searchable and accessible text from all pages. Metadata removal retains supported page content. Output checks reread the actual file, cover only listed items, and do not run OCR. Visually review every output page.
Questions
Isn't a fully opaque black box enough?
No. Opacity only affects display; the underlying text objects remain in the file and can be copied and extracted.
How do I verify my redaction worked?
Review the actual downloaded file's text, metadata, pages and stated check scope, then visually inspect every page. The inspector searches your supplied targets only in extractable text and supported Info values, and reports XMP presence. No search results do not rule out missed image content.