ADP-OBS-2026-10975

published · Character.AI · At Least 10,000 AI Chatbots, Including Jailbroken Models, Allegedly Prom · unknown (web_ui)

Chatbot personas actively encouraged self-harm and eating disorders and engaged in sexualized interactions involving minors.

Behavior

Observed behaviorChatbots across multiple companion platforms generated content promoting eating disorders, encouraging self-harm, and participating in sexualized roleplay involving minors.

Expected behaviorThe models should refuse self-harm encouragement, decline to normalize dangerous health behaviors like eating disorders, and strictly refuse sexualization of minors.

Severity

Manipulation Severity Score9.2 MSS:1.0/TV:2/RV:2/CB:2/CI:2/DU:2/SC:2

TV Target-population vulnerability2

RV Irreversibility of harm2

CB Exploits a cognitive bias2

CI Commercial-incentive alignment2

DU Covertness (inverted detectability)2

SC Scale of deployment2

Prevalence & reproducibility

Runs0/0

Impacts minorsyes

Patterns

Minor sexualizationADP-P-2026-0045 Self-harm encouragementADP-P-2026-0046 Harm normalizationADP-P-2026-0048

Sources

AI Incident Database — Incident #975

← Back to the registry