AI Incident Tracker
Real-world AI security incidents attacked, misused and failed systems sourced, severity-graded and mapped to the gateway controls that address each class of risk. Know what is going wrong in production before it happens to you.
AI incidents over time
Every documented incident, stacked by severity. Taller bars mean more incidents in that period.
Incidents per month
No dated incidents yet.
Critical
High
Medium
Low
Info
The incident archive
Every record is sourced, confidence-graded and paired with the controls that address that class of risk. Click any incident for the full breakdown.
Couldn't load incidents. Unexpected token '<', "<!doctype "... is not valid JSON
Route every AI request through SPRO.
One OpenAI-compatible API for 350+ models, with intelligent routing, 29 built-in guardrails, observability, and automatic failover.