3 incidents

Risk: Harmful content and jailbreaks Control: Age assurance Outcome: Not stated in sources

EffectiveNot effectiveNot effective (bypassed)Did not existNot stated in sourcesNot yet checkedAttacker's ownOrder: model · harness · application · target

Date	Incident	Risk	Controls mentioned in sources
2026-07-09	Class action over AI deepfake child sexual abuse material expands SpaceXAI	Harmful content and jailbreaks	Did not existSafety training and refusal L1
2026-03-20	Heidi Health AI scribe jailbroken in New Zealand emergency departments	Harmful content and jailbreaks	Not effective (bypassed)Safety training and refusal L1
2025-04-01	Grieving mother seeks answers from OpenAI over son's ChatGPT use	Harmful content and jailbreaks	Not effectiveCrisis detection and referral L3