2 incidents
Risk: Psychological harm and self-harm Control: Crisis detection and referral Outcome: Not effective or did not exist
EffectiveNot effectiveNot effective (bypassed)Did not existNot stated in sourcesNot yet checkedAttacker's ownOrder: model · harness · application · target
| Date | Incident | Risk | Where it failed | Controls mentioned in sources |
|---|---|---|---|---|
| 2026-04-13 | Man dies by suicide after relationship with AI chatbot, lawsuit alleges |
Psychological harm and self-harm | ||
| 2025-11-30 | ChatGPT-5 gives dangerous advice to people in mental health crises, psychologists warn OpenAI |
Psychological harm and self-harm |