AI Security & Risk Resources: A Must-Know List
For anyone working in AI governance, risk, or security these resources are essential for understanding how to evaluate and secure AI/ML systems.
Testing & Evaluation / Threat Modeling
1) OWASP ML Security Top 10 – Practical threat modeling for ML systems.
2) NIST AI 100-2 – A foundational taxonomy of adversarial attacks & mitigations.
3) NIST ARIA – Framework for assessing risks and impacts of AI.
Repositories & Incident Databases
1) MIT AI Risk Repository – Curated insights on emerging AI risk areas (recently updated and attached).
2) AI Incident Database (AIID) – Real-world case studies of AI failures and harms.
3) AIAAIC Repository – Broad coverage of algorithmic & automation incidents.
4) OECD Catalogue of Tools & Metrics – Toolkit for assessing trustworthiness of AI.
What resources are you using to evaluate AI risks?