Etoolio

PII Redactor for AI

Free PII redactor. Paste a customer email, ticket or chat and it removes personal data (emails, phones, cards, IBANs, IDs, names) with placeholders, so it is safe to paste into ChatGPT. Runs in your browser.

Use AI at work without leaking customer data

AI assistants are brilliant at drafting replies, summarising tickets and rewording messages. The catch: to help, they need the text, and that text is often full of a customer's personal data. Pasting it into a consumer AI tool can breach data-protection rules and your own promises. This free PII redactor fixes that. It strips the personal data out, gives you safe placeholder text to paste into ChatGPT or Claude, and lets you restore the reply afterwards, all without anything leaving your browser.

What it removes

  • High-confidence data: email addresses, phone numbers, payment cards (checked with the Luhn algorithm), IBANs, IP addresses, reference and account IDs, and API keys or tokens.
  • Names (optional): names that follow greetings and sign-offs like Hi, Dear and Regards.
  • URLs (optional): links, if you want them masked too.

Every distinct value becomes a stable placeholder such as [EMAIL_1] or [NAME_1], and the same value always maps to the same placeholder, so the text still reads naturally and the AI can follow it.

How the workflow goes

  • Paste the customer email, ticket or chat.
  • Copy the safe-to-paste text and use it with your AI tool.
  • Restore the reply: paste the AI's response, still containing the placeholders, into the restore box, and the tool swaps the real values back in locally.

Why this matters

Support and reception teams paste customer data into AI tools every day, usually without meaning to break any rules. A redactor is the bridge that makes AI safe to use at work: staff still get the productivity, but the sensitive data stays on the device. Compliance and security teams can point people to a tool like this instead of trying to ban AI outright.

Honest limitations

Detection is pattern-based and best-effort. It reliably catches structured data like emails, cards and IBANs, but names and free-form postal addresses are genuinely hard, so it can miss some or over-match others. Treat it as a strong safety net, not a guarantee, and always read the safe-to-paste text before you use it. Nothing you paste is ever uploaded: detection, redaction and restoring all run in your browser.

Good to know

PII Redactor for AI, frequently asked questions

What does the PII redactor do?

It scans text you paste and replaces personal and sensitive data, emails, phone numbers, payment card numbers, IBANs, IP addresses, order and account IDs, API keys and names, with neutral placeholders like [EMAIL_1] and [NAME_1]. The result is safe to paste into an AI tool like ChatGPT or Claude. It also gives you a map to restore the AI's reply back to the real values, all in your browser.

Why do I need to redact data before using ChatGPT?

Whatever you paste into a consumer AI tool may be stored by the provider and, on some tiers, used to improve their models. Pasting a customer's email, card number or personal details can breach data-protection rules (like GDPR) and your own privacy commitments. Redacting first lets your team use AI to draft replies and summaries without leaking real customer data.

What types of data does it detect?

High-confidence patterns: email addresses, phone numbers, payment cards (validated with the Luhn check to cut false positives), IBANs, IPv4 addresses, long reference or account IDs, and API keys or tokens. Optionally it also redacts names that follow greetings and sign-offs (Hi, Dear, Regards) and URLs. Repeated values get the same placeholder so the text still makes sense.

Is it accurate? Can it miss something?

It is pattern-based and best-effort, so it can miss unusual formats or names it cannot recognise, and it can occasionally over-match. It is a strong safety net, not a guarantee. Always read the safe-to-paste text before you use it, especially for names and free-form addresses, which are the hardest to detect reliably.

How do I get the AI's answer back to normal?

Keep the placeholder map (it stays in your browser). When the AI replies using the placeholders, paste its reply into the restore box and the tool swaps [NAME_1], [CARD_1] and the rest back to the real values locally, so your teammate or customer sees a normal, complete message.

Is my data uploaded anywhere?

No. Detection, redaction and restoring all run entirely in your browser using JavaScript. Nothing you paste, and none of the mappings, is ever sent to a server. That is the whole point: it keeps the sensitive data on your device.

Is it free?

Yes, free with no signup, and it runs entirely in your browser.

Keep going

Related tools

llms.txt GeneratorGenerate an llms.txt file for your site so AI models can read it.AI Crawler Access CheckerCheck if AI crawlers like GPTBot and ClaudeBot can access your site.AGENTS.md GeneratorGenerate an AGENTS.md and CLAUDE.md for your repo from its GitHub URL.Cold Email Deliverability CheckerCheck a cold email for spam triggers and verify SPF, DKIM & DMARC on your domain.