How to meet it
OWASP AISVS v1.01
Adversarial Robustness · Model Runtime Anomaly Detection
Protect the safety feedback loop
Verify that the safety violation feedback pipeline includes poisoning detection and human review gates to prevent adversarial manipulation of the improvement mechanism.
OWASP AISVS v1.01, requirement 11.4.3
In plain English
If you improve the model from reports of safety problems, check those reports for poisoning and have a person review them before they change the model.
What good looks like
- Feedback data is checked for poisoning patterns.
- A person approves feedback before it is used for training.
- Sources of feedback are tracked.
How teams usually do it
- The poisoning checks from 1.3.1 applied to feedback data, plus a review queue.
Evidence an auditor would accept
- Pipeline configuration and review records.
Common mistakes
- Training directly on user thumbs-up and thumbs-down data.
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.