AI Incidents Dashboard

Tracking reported AI incidents and harmful outcomes from deployed systems

Last Updated: 2026-02-01
Incident Overview
354
Total Incidents Tracked
15
As Deployer
199
As Developer
43
Harmed Parties
Most Incidents:OpenAI
Total Incident Responses:33
About This Data

Description

Tracking AI incidents and harmful outcomes from deployed systems

Note

These incidents represent documented cases from the AI Incident Database. They reflect both direct harm from AI systems and incidents where AI systems were implicated.

Incidents by Organization
Total incidents, breakdown by role (Deployer vs Developer)

Quick Reference

OpenAI
61incidents
🔧 Dev: 75
⚙️ Deploy: 0
Google
58incidents
🔧 Dev: 21
⚙️ Deploy: 0
Facebook
52incidents
🔧 Dev: 1
⚙️ Deploy: 2
Tesla
43incidents
🔧 Dev: 5
⚙️ Deploy: 0
Meta
39incidents
🔧 Dev: 7
⚙️ Deploy: 2
Microsoft
28incidents
🔧 Dev: 9
⚙️ Deploy: 1
Amazon
24incidents
🔧 Dev: 4
⚙️ Deploy: 2
Unknown
19incidents
🔧 Dev: 70
⚙️ Deploy: 8
xAI
16incidents
🔧 Dev: 5
⚙️ Deploy: 0
Apple
14incidents
🔧 Dev: 2
⚙️ Deploy: 0
Key Insights

Most Incidents (Total)

OpenAI leads with 61 documented incidents, followed by Google with 58.

Developer vs Deployer

Incidents where organizations acted as developers (199 total) significantly outnumber those where they deployed systems (15 total).

Scope of Impact

Across all tracked organizations, 43 parties were directly harmed, with 33 documented incident responses.

Understanding the Data

Deployer vs Developer

Deployer: Organization deployed an AI system in production.
Developer: Organization developed the AI model/system.

Harmed By

Number of distinct entities that were harmed by incidents involving this organization's AI systems.

Related Entities

Other organizations, platforms, or systems that were mentioned or implicated in incidents.