Shared catalogs of AI threats give you a common vocabulary for threat models, test plans, and findings. Each one covers different ground, and your auditors, customers, and engineers will cite different ones. Learn which is which and how they map to each other.
Read in this order
- OfficialOWASP AI ExchangeStart here: a broad, practical map of AI threats and the controls for each.
- OfficialSAIF: Google's Guide to Secure AIGoogle's view of secure AI, useful for seeing how a vendor frames the same risks.
- OfficialState of Agentic AI Security and Governance (OWASP GenAI Security Project)Where agent-specific security and governance work stands now, and who is publishing what.