Home/Incidents/Multi-Category Severity

These visualizations are experimental and still being refined. Designs and data presentation may change.

Multi-Category Severity

Incidents that score high across multiple harm categories simultaneously represent especially damaging events. This visualization filters incidents by their summed direct severity score across all 10 harm categories, highlighting those causing significant harm in more than one dimension.

How to read this

Each bar represents a year, stacked by risk domain. The chart shows incidents whose summed direct severity score across all 10 harm categories exceeds the selected threshold (moderate >10 or high >15). A high summed score indicates incidents that caused significant harm in more than one category. Toggle between count (absolute numbers) and proportion (relative share per year). Click any bar segment to view matching incidents.

Key Takeaways

1.1,053 incidents have a summed direct severity score exceeding 10 across the 10 harm categories.
2.Malicious Actors & Misuse accounts for the largest share of multi-category severe incidents at 43% (450 incidents).
3.Multi-category severe incidents decreased by 94% from 2025 to 2026 (306 to 19).
4.Socioeconomic & Environmental showed the largest absolute increase over the last 3 years, growing by 1 incidents in multi-category severity.

Discrimination & Toxicity

Privacy & Security

Misinformation

Malicious Actors & Misuse

Human-Computer Interaction

Socioeconomic & Environmental

AI System Safety, Failures & Limitations

Source data from the AI Incident Database and MIT AI Risk Repository. Incidents prior to 2015 not shown.