Statistics
The index in numbers
Auto-computed from the live corpus, as of June 16, 2026. Every number carries a copy-ready citation; each row has a stable anchor. Full dataset on the open data page.
The AI Failure Index documents 532 AI failures in production, citing 1379 sources across 11 industries.
Hallucination accounts for 35 percent of indexed production failures (188 of 532).
Failure-mode share: Hallucination 35%, Policy Violation 23%, Agentic Action Error 12%, Brand & Safety Incident 12%, Prompt Injection 7%, Data Leakage 5%, Tool Misuse 4%, Identity & Access Drift 2%.
Severity distribution: Catastrophic 33, High 277, Medium 196, Low 26.
Most-affected industries: Public Sector (120), Cross-industry (114), SaaS (111), Legal Services (66), Retail & E-commerce (35).
75 percent of entries are generative-AI failures (400 of 532); the rest are predictive systems: vision, scoring, recommenders, and autonomous systems.
Entries indexed per month: 2026-06: 507, 2026-05: 25.
Method notes: counts cover live entries only (merged duplicates excluded); severity follows the rubric on the methodology page; hallucination and the other seven modes are defined in the versioned taxonomy.