Home › AI Risk Index › Incidents › 2026

AI Incidents 2026

2 recorded · permanent

Every AI headline our engine weighed at 55 or more during 2026: harm done, attempted or narrowly avoided. Numbered in the order they were recorded, never edited, never removed.

September 2026

2 incidents

#1 · Uncontrolled AI could lead to 'silicon species' rivalling humans, warns Microsoft

18 September 2026 · BBC Business Safety incidentsCapability jumps

59 #2 · OpenAI ‘ethically hacked’ with help of Anthropic’s Claude chatbot

18 September 2026 · Guardian Biz Regulation & controlMisuse & cyberCapability jumps United States

All years · ai_incidents.json · CSV · CC BY 4.0