Lion Cipher Lab
Detect and redact PII with Microsoft Presidio
- AI Security
- Beginner
- ~40 min
- Free · no account required
Why it matters
The advice "don't feed secrets to AI" is easy to say and hard to actually do -- the moment you are pasting a support ticket, a log line, or a spreadsheet row into a model, the sensitive bits are buried in the middle of text you do want to send. PII detection is the control that closes that gap: instead of trusting yourself to spot every name, email, and card number by eye, you run the text through a tool that finds those entities for you and blanks them out before the data goes anywhere. Microsoft Presidio is a widely used open-source version of exactly this, and learning it turns a good…
What you'll learn
- PII detection
- data anonymization and redaction
- using Microsoft Presidio
- sanitizing data before it reaches an LLM
- recognizing PII entity types
- building a defensive data pipeline habit
Steps in this lab
- The control behind "don't feed secrets to AI"
- Meet Presidio: an analyzer and an anonymizer
- Install Presidio
- Write a sample text you own
- Analyze the text for PII entities
- Know what the entity labels mean
- Redact and anonymize the text
- Where this fits -- and the instinct to keep