Guide · Security & privacy
How to find and redact personal data in a PDF
Reviewed
Short answer
To find and redact personal data in a PDF, run a detector over the text to list candidates — emails, phone numbers, ID and card numbers, IBANs and, with AI, names and addresses — review each one, then apply true redaction that removes the text rather than covering it, and check the output no longer contains it. In PDFBetter, AI redaction review does all four steps; pattern checks are free and AI categories cost 1 AI credit per page.
On this page (5)
Step by step
- 1
Open AI redaction review
Upload the PDF or pick a document from your account. A free account is needed. Unlock password-protected files first.
- 2
Choose what to look for
Free pattern checks find emails, phones, URLs, US Social Security numbers, IBANs, card numbers, Turkish TC Kimlik numbers and dates of birth. Add AI categories for names, postal addresses, organizations and other details.
- 3
Review the suggestions
On the page, black boxes are selected for redaction and red outlines are suggested but not selected. Tick or untick each item, or use Select all and Clear all.
- 4
Apply
Selected items become redaction marks, and pages with marks are rebuilt without the underlying text.
- 5
Read the check
The result is re-read. Anything listed as still present must be marked by hand in the editor and applied again.
1Sensitive text
2Mark what to remove
3Removed from the file
What the detectors find, and miss
Numbers with check digits — IBANs, card numbers, TC Kimlik numbers — are validated, so those matches start ticked. AI suggestions that aren't high-confidence start unticked, so nothing uncertain is removed by default.
Suggestions can miss handwriting, text inside images and stamps, names split across lines and unusual formats. Pages without a text layer can't be searched until OCR adds one. Read the whole document before you share it, and mark anything missed with the manual Redact PDF tool.
Real redaction, not a black box
Drawing a black rectangle or highlighting text in black leaves the words in the file; anyone can copy them out. Applied redaction rebuilds the marked pages as images without the original text, which is why the remaining text on those pages is no longer selectable or searchable afterwards.
Tools used in this guide
- AI redaction reviewUpload a PDF, review suggested sensitive text by category, apply real redaction to what you select, and see a check that it is gone from the text.
- Redact PDFMark what must go, apply redaction, and download a PDF where the marked content no longer exists in the file.
- OCR PDFUpload a scan, pick its language, and download a PDF that looks the same but lets you search, select and copy the text.
- Protect PDFSet a password to open the file, optionally restrict printing, editing and copying, and download an encrypted PDF.
Questions people ask
Does it redact automatically?
No. It only suggests. You decide what to apply, and uncertain AI suggestions are never pre-selected.
Is the file sent to an AI provider?
Only if you choose AI categories; then page text is sent to the AI provider under no-training terms. Pattern-only checks run on PDFBetter's servers.
How is this different from the manual Redact PDF tool?
Redact PDF is where you draw boxes yourself. AI redaction review finds candidates first and then applies the same true redaction. Use the manual tool for images, signatures and anything missed.
Does redaction remove data from images in the PDF?
Only where you mark the image area. The text check can't read text inside images, so look at photos and scanned stamps yourself.
Can I redact many PDFs at once?
Not in one step. A workflow can run redaction suggestions and a review on each document, but applying redaction stays a manual action per document.