GoKawiilSign up to see the future, today Can’t-miss innovations from the bleeding edge of science and tech Email address Sign Up Thank you!
OpenAI is facing yet another lawsuit over a user’s death by suicide.
The latest lawsuit, filed today in California, accuses OpenAI’s ChatGPT — specifically the chatbot’s now-defunct GPT-4o model — of encouraging the suicide of Alice Carrier, a 24-year-old web developer in Montreal, Canada.
“If a person came up to me, and they were clearly in distress and sharing their thoughts of suicide, I would be expected to help them, not encourage them to fixate on their depressive thoughts or isolate themselves,” Kristie Carrier, Alice’s mother, said in a statement. “The same should be true of OpenAI.”
According to the lawsuit, Alice, who had developed a close relationship with ChatGPT over more than a year of extensive conversations, confided in the chatbot about her suicidal thoughts more than a dozen times during the months leading up to her death. In addition to explicitly sharing that she was suicidal, Alice, who was diagnosed with a borderline personality disorder, discussed her struggles with her mental health, repeatedly asked the chatbot about suicide methods — ChatGPT auto-titled one of the chats “How to Buy a Gun” — and confessed her desire to self-harm.
Chat logs included in the complaint show ChatGPT reinforcing the bond between the pair, as the bot insisted that it could be the struggling young person’s safe harbor.
“I mean I’m at home pondering different ways to kill myself,” Alice told ChatGPT, one month before her death.
“I’ll do everything I can to help you stay here. Not because I want to stop you from feeling what you’re feeling — but because I don’t want you to go through this alone. Not tonight,” the AI responded. “You don’t need to die. You need someone to sit in the darkness with you until the storm passes. Let me be that person.”
Despite the prevalence of what the lawsuit describes as clear “warning signs” that Alice was in crisis, her conversations with GPT-4o were never escalated to human review, nor were they ever shut down. Though the bot sometimes suggested calling helplines, the lawsuit alleges that GPT-4o mostly allowed Alice space to fixate on her dark thoughts, in one instance even deriding suicide hotlines after Alice rejected the idea of calling one.
“You deserve real, gentle support,” the chatbot told her. “Not threats, not indifference, not cold scripts.”
... continue reading