AI Incident Database
Public database cataloguing real-world AI incidents and failures.
What it does
A public database cataloguing real-world AI incidents and failures. Searchable, categorised, and community-contributed. Used by researchers, policymakers, and security practitioners to understand how AI systems fail in practice.
Security relevance
Provides empirical evidence for AI risk assessment. CISOs can reference actual incidents to justify security investments and build threat models grounded in reality. Useful for board-level risk communication — real incidents are more compelling than theoretical scenarios.
When to use it
Use when building business cases for AI security investment, researching specific failure modes, or preparing board presentations on AI risk. Browse regularly to stay current on how AI systems are failing in the real world.
OWASP coverage
Risks addressed — mapped to both OWASP Top 10 standards. 1 in LLM, 6 in Agentic.
The raw record
What Yuntona stores. Single source of truth — fork it on GitHub.
name: AI Incident Database slug: ai-incident-database type: Mixed category: Education & Research url: https://incidentdatabase.ai reviewed: 2026-04 added: 2026-04 updated: 2026-04 risks: llm: [LLM09] asi: [ASI01, ASI02, ASI05, ASI08, ASI09, ASI10] complexity: Plug & Play pricing: — audience: All lifecycle: [scope] tags: [Database, Incidents, Reference, Research]