Privacy redactor: hide e-mails, phones, IDs, IBANs and API keys from text
Before pasting a document into an AI chat or sharing a conversation, replace personal data with labels or consistent aliases like [EMAIL_001]. Detection happens on this page, never on a server.
Ready when you are.
What it detects
- E-mail addresses and phone numbers (international and Spanish formats).
- Spanish DNI and NIE (validated with the control letter), IBAN (validated with its checksum) and payment card numbers (Luhn check).
- IPv4/IPv6 addresses, and secrets: OpenAI, Anthropic, GitHub, AWS, Google, Stripe and Slack keys, JSON Web Tokens, private keys, database URLs with passwords and password assignments.
- Names that follow greetings, signatures or “my name is…” (lower confidence — check them).
Know the limits
- Detection is pattern-based: it can miss data (addresses, unusual formats, names without cues) or flag false positives. Always read the result.
- Aliases are pseudonymization, not anonymization: context can still identify a person.
- For whole conversations with consistent aliases across many messages, review per value and an exact preview, use AI Vault’s Share safely.
What is the difference between labels and aliases?
Labels replace every value with a generic tag such as [REDACTED EMAIL]. Aliases give each distinct value its own consistent tag ([EMAIL_001], [EMAIL_002]), so the text still makes sense; secrets are always removed, never aliased.
Is this enough to comply with GDPR?
No tool can guarantee compliance on its own. This helps you remove obvious identifiers; you remain responsible for what you share and for checking the result.
Does my text leave the browser?
No. The redactor runs in JavaScript on this page and makes no network requests with your text.
Need to clean whole conversations?
AI Vault imports your ChatGPT, Claude and Gemini exports, finds personal data across every message and exports exactly what you choose.
Open AI Vault