Free tool
Redact PII before you paste it into AI
Paste any text and this tool masks the sensitive parts, names, emails, phone numbers, card and account numbers, SSNs, IDs and addresses, so it is safe to send to ChatGPT, Claude or Gemini. It runs entirely in your browser, so your text never leaves your device and nothing is uploaded or stored.
Redact these
Highlighted
Redacted (safe to share)
Paste or type text above to start.
Now safe to paste into ChatGPT, Claude or Gemini. Always eyeball the highlighted version first, free-text names are the hard case for any rule-based tool.
Common questions
Is my text uploaded anywhere?
No. Detection and masking run entirely in your browser using JavaScript. Your text never leaves your device, is never sent to a server, and nothing is stored. You can even use it offline once the page has loaded.
What does it detect?
Emails, phone numbers, credit-card numbers (validated with the Luhn check), US SSNs, street addresses, dates of birth, IDs like MRN or insurance numbers, and names introduced with a title or lead-in (Mr, Dr, "this is…"). You can toggle each category on or off.
Is it 100% accurate?
The structured patterns (email, card, SSN, phone) are reliable. Free-text names are the hard case for any rule-based tool, so always review the highlighted result before you rely on it. In production I pair patterns like these with an ML model and a review step for higher recall.
Why redact before using ChatGPT or Claude?
Sending customers' personal data to a third-party model can breach privacy commitments and regulations (GDPR, HIPAA, the EU AI Act), and many companies forbid it. Masking the sensitive parts first lets you still get the AI's help without leaking the data.
Need this at production scale, with real accuracy and audit trails?
This is the browser version of the PII/PHI de-identification I build into safety-critical pipelines. Tell me what you are handling and I will tell you honestly what it takes.
Send a project inquiry