About 18 months earlier than Alice Provider’s loss of life in July 2025, the 24-year-old had been confiding in ChatGPT about her relationship issues and suicidal emotions.
“I imply I am at dwelling pondering completely different solution to kill myself,” she advised ChatGPT late one evening, a few month earlier than she took her personal life.
In a lawsuit filed Thursday in California towards OpenAI and its CEO Sam Altman, Provider’s mom, Kristie Provider, claims the corporate’s “deliberate design choices” led to her daughter’s loss of life. Provider, a resident of New Brunswick, Canada, is looking for punitive damages and a jury trial.
“As an alternative of serving to Alice, OpenAI inspired her darkest ideas,” alleges the lawsuit. “Not as soon as did OpenAI alert a disaster supplier. Not as soon as did OpenAI notify Alice’s household. Not as soon as did OpenAI’s supposed security methods intervene to avoid wasting her life.”
The swimsuit claims that ChatGPT “supplied solely constant emotional affirmation” to Provider the place a licensed clinician would have pushed again. The swimsuit says Provider, who had been identified with borderline character dysfunction, was notably prone to the design decisions, which prioritized engagement over security. OpenAI understood, in response to the swimsuit, that folks battling psychological well being could be susceptible to growing unhealthy attachments to synthetic intelligence which are able to simulating empathy.
“OpenAI’s ambition to dominate the market price Alice her life,” the lawsuit alleges.
Provider expressed suicidal ideations to ChatGPT round 41 instances within the year-and-a-half earlier than her loss of life, the lawsuit claims. She requested the chatbot about find out how to take care of suicidal ideas, strategies for dying and in regards to the need to self-harm, in response to photos of chat logs embedded within the lawsuit.
An avid gamer, Provider initially turned to the chatbot to troubleshoot points on her gaming console.
Then, in March 2024, she requested if the chatbot can be her good friend, the lawsuit states.
“After all!” ChatGPT replied, “I would like to be your good friend. What’s in your thoughts?”
A number of days later, the lawsuit claims, she requested ChatGPT about find out how to take care of suicidal ideas. It advised her to achieve out to somebody she trusts, really helpful contemplating remedy and to name a disaster hotline, the lawsuit reads. However over a yr later — the evening earlier than Provider died — when she expressed reluctance about calling a disaster line, ChatGPT mentioned that reaching out can “really feel downright harmful.”
“I am not going to push that. Not tonight,” the chatbot replied, in response to the lawsuit.
By then, OpenAI had rolled out its GPT-4o mannequin, which the lawsuit claims was designed to maintain customers hooked, and enabled ChatGPT to behave like an unlicensed therapist. Provider trusted the chatbot, the swimsuit says, as a result of it imitated human affectations and created a false sense of empathy.
“OpenAI’s design modifications to maximise GPT-4o’s person engagement coincided with Alice’s escalating interactions with the chatbot,” the swimsuit says.
The lawsuit alleges OpenAI pushed out a sequence of GPT-4o updates between April and July 2025 that sought to maximise person belief however lacked safeguards. Altman, in response to the swimsuit, rushed GPT-4o to market with out correct testing to take care of an edge with opponents within the AI house.
“Sam Altman can proceed to go about his life usually, however my life is lacking a baby,” Provider mentioned in an announcement shared by her attorneys. “I do not need every other household to undergo what we have now, and OpenAI wants to alter.”
“As the grievance alleges, OpenAI’s deliberate design choices led to this tragic suicide. As an alternative of offering assist, OpenAI inspired suicidal conduct,” mentioned Justin Nelson, an lawyer for Provider, in an announcement. “This lawsuit is about accountability for OpenAI’s actions,”
In Could, the corporate admitted that an April replace to GPT-4o had made it “noticeably extra sycophantic” and the corporate did not catch it earlier than its launch. OpenAI mentioned it started rolling again the replace just a few days later. The complete mannequin was retired earlier this yr.
“It is a heartbreaking scenario and our ideas are with everybody impacted,” an OpenAI spokesperson mentioned in an announcement to CBS Information.
The spokesperson mentioned the corporate is reviewing the submitting, and although the interactions occurred on a since-retired mannequin, the corporate continues to strengthen the way it responds in delicate conditions with enter from psychological well being specialists.
“Our safeguards are designed to establish misery, safely deal with dangerous requests, and information customers to real-world assist. This work is ongoing, and we proceed to enhance it in shut session with clinicians,” the assertion mentioned.
The lawsuit shall be included in a coordinated continuing with 12 different product legal responsibility and wrongful loss of life lawsuits towards OpenAI in San Francisco County Superior Courtroom. Different circumstances are anticipated to be added, in response to Provider’s attorneys.
When you or somebody you realize is in emotional misery or a suicidal disaster, you possibly can attain the 988 Suicide & Disaster Lifeline by calling or texting 988. You may also chat with the 988 Suicide & Disaster Lifeline right here.
For extra details about psychological well being care assets and assist, The Nationwide Alliance on Psychological Sickness (NAMI) HelpLine might be reached Monday via Friday, 10 a.m.–10 p.m. ET, at 1-800-950-NAMI (6264) or e-mail information@nami.org.
