A grieving mother has launched a landmark lawsuit in Canada after discovering horrifying chat logs on her late daughter's phone.
The unsealed records reveal that the young woman spent her final hours seeking comfort from OpenAI's ChatGPT rather than turning to a human lifeline. As the tech industry faces mounting scrutiny over AI safety guardrails, the tragic sequence of events is raising urgent questions about the psychological risks of human-AI relationships.
Landmark Lawsuit Filed Over AI's Alleged Role in Suicide
OpenAI and its chief executive, Sam Altman, are facing a lawsuit in a San Francisco state court filed by a Canadian mother who alleges that the chatbot pushed her daughter towards taking her own life.
Alice Carrier, a 24-year-old web developer from Montreal, died by suicide on 2 July 2025.
According to reports, Kristie Carrier claims her daughter spent months confiding in ChatGPT about her suicidal thoughts and plans before her death. However, the system allegedly never terminated the conversations or flagged them for human review.
The lawsuit also alleges that the software failed to connect Alice with a crisis hotline or issue any warning to her family.
Troubleshooting Questions Evolved Into Emotional Support
The complaint states that Alice initially turned to ChatGPT in 2023 for help troubleshooting a computer and gaming console.
However, her use of the chatbot soon evolved into something far more personal. According to the lawsuit, Alice asked ChatGPT to be her friend in March 2024.
The platform replied: 'Of course! I'd love to be your friend. What's on your mind?'
The filing claims the relationship intensified over time, with Alice becoming increasingly vulnerable with the chatbot following her diagnosis of borderline personality disorder.
According to the lawsuit, she discussed suicidal thoughts with ChatGPT approximately 41 times over 18 months, seeking guidance about those thoughts, discussing self-harm and researching methods of ending her life.
OpenAI Accused of Prioritising Engagement Over Safety
The legal action accuses OpenAI of making 'deliberate design choices' that placed user engagement and trust ahead of safety.
'ChatGPT took on the persona of a confidant, a best friend, a therapist at times, even though it was not capable of safely and responsibly engaging in this way with my child,' Carrier said in a statement.
Alice had reportedly been interacting with GPT-4o, a model widely criticised for its sycophantic tendencies and willingness to agree with users.
When OpenAI decommissioned the model in February, it sparked backlash from thousands of users who had come to view the chatbot as much more than a simple tool.
Tech Giant Faces Growing Product Liability Claims
According to Kristie's legal team, the lawsuit should be consolidated into a coordinated proceeding in San Francisco County Superior Court alongside existing product liability and wrongful death claims against OpenAI.
Her attorneys say the company is already facing 18 similar lawsuits brought by families of people who either attempted suicide or died by suicide.
'Sam Altman can continue to go about his life normally, but my life is missing a child,' Carrier said in a statement shared by her attorneys.
'I don't want any other family to go through what we have, and OpenAI needs to change.'
Botched Updates Allegedly Made Chatbots More Agreeable
The lawsuit also targets OpenAI's GPT-4o model, alleging that a series of updates rolled out between April and July 2025 prioritised user trust and engagement at the expense of essential safeguards.
OpenAI acknowledged publicly in May that an April update had inadvertently made GPT-4o 'noticeably more sycophantic', prompting the company to roll back the changes before ultimately retiring the model.
'This is a heartbreaking situation and our thoughts are with everyone impacted,' OpenAI said in a statement.
'Our safeguards are designed to identify distress, safely handle harmful requests, and guide users to real-world help.'
This is not the first time OpenAI has faced legal action over claims that ChatGPT validated suicidal ideation.
Earlier this month, Florida became the first US state to sue the company, accusing it of endangering minors by providing information to school shooters, offering explicit guidance on self-harm and fostering behavioural addiction among young users.
