Is the chat log uploaded anywhere?

No. Anonymization runs entirely in your browser with pattern matching. The raw transcript never leaves your device, which is what makes it safe to paste sensitive logs.

Why use consistent pseudonyms?

Replacing the same email or name with the same placeholder everywhere keeps the conversation coherent. A reader can still follow who said what without seeing any real identity, which matters for debugging and research.

How reliable is name detection?

It is heuristic — it matches capitalised two-word names, which catches most full names but can miss single names or over-match unusual capitalisation. Always review the output before sharing.

What is the difference between this and the prompt sanitizer?

This tool is built for multi-turn transcripts and pseudonymizes consistently across the whole conversation, including person names. The prompt sanitizer focuses on bulk-cleaning a library of standalone prompts.

Can I share anonymized logs as a dataset?

Anonymization greatly reduces risk, but true de-identification for public datasets is a higher bar — combinations of details can still re-identify someone. Review carefully and follow your organisation's data-handling policy.

AI Chat Log Anonymizer

AI chat log anonymizer

Real conversation logs are gold for debugging a flaky assistant or studying how users phrase things — but they’re full of names, emails, order numbers, and other details you can’t paste into a public bug tracker or research dataset. This tool anonymizes a transcript in your browser, replacing identifiable information with consistent pseudonyms so the conversation stays readable and useful without exposing anyone.

How it works

Paste the chat log and toggle which categories to anonymize: emails, phone numbers, URLs, IP addresses, card numbers, account or ticket references, and person names. Each category uses consistent pseudonyms — the same email always becomes [EMAIL_1], the same person always becomes [PERSON_2] — so the back-and-forth still makes sense to whoever reads it. Live counts show how much was replaced, and everything runs locally so the raw, sensitive transcript never leaves your machine.

Tips and notes

Local only. The raw log is never uploaded — paste freely.
Consistency aids debugging. Because identical values map to identical placeholders, a reader can still trace the conversation logic.
Name detection is heuristic. It catches most full names but review the output for single names, nicknames, or unusual formatting it may miss.
Datasets need more care. For public release, remember that combinations of non-obvious details can still re-identify someone — review beyond the automated pass.