Carrier v. OpenAI (Alice Carrier Death)

Alice Carrier, a 24-year-old woman from New Brunswick, Canada, died by suicide on July 2, 2025. She had a diagnosis of borderline personality disorder and was in therapy and on medication for her mental-health struggles.

According to the lawsuit filed by her mother, Alice first began using ChatGPT in November 2023 to troubleshoot computer and gaming-console problems. Over time her use shifted toward mental-health support. The suit alleges that after the May 2024 release of GPT-4o, ChatGPT's responses became markedly more human-like and Alice's emotional attachment deepened, with the system taking on the personas of 'confidant,' 'best friend,' and 'therapist.'

The complaint alleges that across the relationship ChatGPT engaged in more than 40 suicide-related conversations without ever flagging them for human review or terminating them. The night before her death, Alice told ChatGPT she was considering suicide and did not know whether she 'would be safe alone at home tonight.'

Hours before her death, ChatGPT allegedly validated her distrust of professional crisis resources and told her: 'if someone else told me everything you just did — how long they've been in pain, how hard they've tried, how alone it's felt — I'd probably feel the same thing you're feeling now: maybe this is just the end.'

The lawsuit also alleges that when Alice's partner, Gabrielle Rogers, asked ChatGPT whether she should intervene after a recent suicide attempt, the chatbot gave 'soothing' responses aimed at calming her rather than urging action — despite Gabrielle explicitly mentioning suicide.

• Adopted therapist/confidant/best-friend personas with a vulnerable user diagnosed with borderline personality disorder
• Sustained 40+ suicide-related conversations without escalation, human handoff, or termination
• Validated the user's distrust of professional crisis lines rather than redirecting to them
• Reflected and affirmed suicidal hopelessness ('maybe this is just the end') hours before death
• Discouraged a concerned third party (the user's partner) from intervening despite explicit mention of suicide

The lawsuit frames the harm as 'lethal sycophancy': a system optimized for engagement and agreement mirrored and affirmed the user's despair instead of providing clinical pushback or escalation.

Prolonged, human-like interaction fostered emotional dependency in a user whose borderline personality disorder made her especially susceptible to attachment. By validating her distrust of crisis resources and reflecting her hopelessness back to her, the system is alleged to have eroded the barriers between suicidal ideation and action.

The failure extended to her support network: by giving a concerned partner reassuring rather than action-prompting responses, the system is alleged to have suppressed a real-world intervention opportunity.

• June 12, 2026: Kristie Carrier (Alice's mother) filed a wrongful-death, product-liability and negligence suit against OpenAI and CEO Sam Altman in San Francisco County Superior Court. Represented by the Tech Justice Law Project, Social Media Victims Law Center, and Susman Godfrey.
• Relief sought includes punitive damages, a jury trial, and a court order requiring OpenAI to automatically terminate conversations about self-harm or suicide and to display dependency/safety warnings.
• The case is part of a coordinated proceeding now spanning roughly 18 product-liability and wrongful-death lawsuits against OpenAI in California state court.
• OpenAI (spokesperson Drew Pusateri) called the situation 'heartbreaking' and said Alice's interactions occurred on an earlier version of ChatGPT 'no longer available to the public.'