Pasting raw text into an AI assistant is convenient, but customer emails, support tickets, and logs are full of personal data you should not be sending to a third party. This PII redactor scans your text in the browser and replaces structured identifiers — emails, phone numbers, card numbers, IP addresses, URLs, dates of birth, and street addresses — with consistent placeholder tokens, so you can get the help you need without leaking anyone’s details.
How it works
The tool runs a set of conservative regular expressions over your text, one per identifier type. Each unique value is assigned a stable token the first time it appears — the first email becomes [EMAIL_1], the second distinct one [EMAIL_2] — and every later occurrence of the same value reuses that token. This preserves relationships in the text (the AI can still see that the same person appears twice) while hiding the real value. A summary lists every match grouped by type so you can confirm nothing important was missed or over-redacted.
Detection runs entirely client-side. There is no network call, no logging, and no storage, so even the original unredacted text never leaves your machine. That makes the tool safe to use on genuinely sensitive material as a first cleaning step.
Tips and examples
Always read the detection summary before copying. Pattern matching cannot find personal names reliably — names have no fixed shape — so this tool deliberately skips them and focuses on identifiers it can match accurately. Scan the output yourself for names, account numbers, and any niche identifiers specific to your domain, and redact those by hand.
Treat redaction as one layer, not the whole answer. For regulated data, pair this tool with human review and keep the consistent placeholders in your prompt so the AI’s response maps cleanly back onto your real data once you re-insert it locally.