Redact personal data from a PDF before you give it to AI
When you upload or paste a document into ChatGPT, Copilot or Gemini, your client's data ends up on a third-party server and you can't pull it back. DocRazor finds and truly removes it from the PDF, entirely in your browser, before the file ever leaves your hands.
Load the PDF and open the Redact tool.
Press 'Find personal data': names, tax codes, IBANs and emails are marked automatically.
Redact and download the cleaned PDF, then hand it to the AI safely.
FAQ
Why redact a PDF before using it with AI?
When you send a document to ChatGPT, Copilot or Gemini, its content is processed on the provider's servers, often abroad, and may be retained. Removing personal data first (names, tax codes, IBANs, emails) avoids transferring client or third-party data to an external service — a real concern under the GDPR.
Is the PDF uploaded anywhere to be redacted?
No. Detection and redaction happen 100% in your browser: the document never leaves your device. You can verify it in the developer tools (F12).
Is the data really removed or just covered?
Really removed. The redacted page is re-rendered as an image, so there is no selectable or copyable text left under the black. That's the key difference from a black rectangle drawn on top.
More local-first PDF tools
Part of the Secure Local Suite — sanitize text, extract mind maps, edit video and optimize PDFs, all in your browser.