How to meet it
OWASP AISVS v1.01
Monitoring, Logging & Anomaly Detection · Detection and Alerting
Custom rules for AI attack patterns
Verify that custom rules detect AI-specific threat patterns for coordinated jailbreak attempts, prompt injection, and system prompt extraction attempts.
OWASP AISVS v1.01, requirement 12.2.3
In plain English
Write custom detection rules for attacks specific to AI, such as coordinated jailbreak attempts and attempts to extract the system prompt.
What good looks like
- Rules cover coordinated jailbreaks, injection and prompt extraction.
- Rules are tested and updated.
- Alerts are reviewed.
How teams usually do it
- SIEM rules using the AI log schema, plus canary strings in system prompts.
Evidence an auditor would accept
- Rule definitions and test results.
Common mistakes
- Relying only on generic security rules that know nothing about AI.
Not sure which requirements apply to your system? Take the assessment.
Requirement text from the OWASP AI Security Verification Standard, licensed under CC BY-SA 4.0. This guidance is shared under the same licence.