How to meet it
OWASP AISVS v1.01
Adversarial Robustness · Membership-Inference and Model-Inversion Mitigation
Rate limits sized against model extraction
Verify that inference endpoints enforce per-principal and global rate limits sized to the extraction threat model, and not solely as a generic API throttle.
OWASP AISVS v1.01, requirement 11.2.2
In plain English
Set rate limits per caller and overall that are low enough to make copying your model through its API slow and costly.
What good looks like
- Limits are set from an estimate of how many queries extraction would take.
- Limits apply per caller and globally.
- Limit breaches are logged and reviewed.
How teams usually do it
- API gateway rate limiting with per-key and per-account quotas.
Evidence an auditor would accept
- The extraction estimate and the configured limits.
Common mistakes
- Generic limits set only to protect capacity.
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.