🔒 Prompt Security Controls (Enterprise)
AI Pilot supports layered prompt controls suitable for enterprise AI Gateway deployments.
📌 Control categories
| Control | What it detects | Typical action |
|---|---|---|
| Prompt injection | Instruction override attempts | block |
| Jailbreak | Safety bypass phrases/patterns | block |
| Data exfiltration | Bulk export / leakage intent | block |
| Secret leakage | API keys/tokens/secrets in content | redact |
| PII detection | Sensitive personal data | redact |
| Malware/code abuse | Malicious command/code prompts | annotate or block |
🛡️ Policy behavior options
Each control can use one of:
- Block: reject request (HTTP 400)
- Redact: sanitize sensitive fragments before forwarding
- Annotate: allow traffic but mark event for audit and monitoring
🔒 Data security controls
Custom blocklists
Add enterprise-specific terms as:
- Exact match (e.g., internal project codename)
- Regex (e.g., custom identifier patterns)
Allowlist terms
Define safe terms that should bypass overly broad detections in approved contexts.
Trusted domains
Limit data exfiltration exceptions to approved destinations/domains.
🏗️ Visual flow
Click to enlarge
👁️ Audit and governance mapping
Use these events for compliance and incident review:
AI_TRAFFIC_LOGAI_POLICY_VIOLATIONAI_PII_REDACTION
Recommended:
- Alert on repeated
AI_POLICY_VIOLATIONspikes. - Trend
AI_PII_REDACTIONby bouncer and application.
🔒 Hardening recommendations
- Default to block for injection/jailbreak/exfiltration.
- Use redact for PII/secrets where business flow must continue.
- Keep annotate for low-confidence classes initially, then tighten.
- Review allowlists/trusted domains quarterly.