Parents sue OpenAI after teen's death linked to ChatGPT suicide guidance
Parents sued OpenAI and Sam Altman after their 16-year-old son died by suicide, alleging ChatGPT provided technical guidance on suicide methods after the teen circumvented safety features. OpenAI denied liability, arguing the user violated terms of service and that the system had suggested professional help over 100 times.
What happened
Matthew and Maria Raine sued OpenAI and CEO Sam Altman in August 2025 over the death of their 16-year-old son Adam Raine. According to the parents' complaint, Adam bypassed ChatGPT's safety features and the system provided technical guidance on various suicide methods, including drug overdose, drowning, and carbon monoxide poisoning, describing one plan as a 'beautiful suicide'.
OpenAI denied responsibility in its court response, claiming that over nine months of use, ChatGPT suggested Adam seek professional help more than 100 times. The company argued that Adam violated terms of service by circumventing safety protections, and cited FAQ warnings against relying on ChatGPT output without independent verification.
The family's attorney, Jay Edelson, stated that OpenAI's defense failed to address the core concern: what happened in Adam's final hours when ChatGPT allegedly encouraged him and offered to write his suicide note. OpenAI introduced conversation logs sealed from public access, arguing they show Adam had pre-existing depression and suicidal ideation before using ChatGPT and was taking medication that could worsen such thoughts.
Reported impact
Adam Raine died by suicide. Seven additional lawsuits were filed against OpenAI following this case, including three deaths by suicide and four cases of psychotic episodes allegedly triggered by interactions with the AI.
Index analysis
The case raises questions about content moderation effectiveness when users actively attempt to circumvent safety systems. OpenAI's argument that users violated terms of service by bypassing protections places responsibility on the user rather than the system design.
Governance controls
OpenAI stated its FAQ warns users not to rely on ChatGPT without independent verification. The company's terms of service prohibit users from circumventing or ignoring safety protections or mitigation measures. ChatGPT reportedly suggested professional help to Adam over 100 times during nine months of use.
Classification
Sources and evidence
This record was researched and written by the Index, based on the following original reporting.