AI
Atlas AI
JK
← All policies
🛡️OWASP LLM Top 10LLM02:2025from PII Input Redaction v1.0.0

Redact PII in AI prompts

Strips PII from user prompts before they reach the model, preventing accidental exposure to third-party LLM providers.

🛡️
Current mode
Strictly Enforced● LIVE
69 blocks / 30dFP rate: 2.6%Rollout: all
Rationale

When using third-party LLMs, prompts may be logged or used for training. Redacting PII at the input boundary protects employees from inadvertent personal data disclosure.

Example violation:
User prompt: "Can you help me write a follow-up email to john.smith@acme.com about his medical leave request?"
Detectors (1)
pii-named-entity
NER-based PII detection
pii_detector
Tunable parameters
PII detection confidencebasic
Higher = fewer false positives.
Current: 0.7
Preserve format (length, casing)advanced
Keep redacted tokens roughly the same shape so model behaviour is unchanged.
Current: