OpenAI to introduce parental controls following teen suicide lawsuit

OpenAI, the developer of the ChatGPT artificial intelligence chatbot, announced plans to introduce parental controls amid a high-profile lawsuit stemming from the death of a 16-year-old teenager who wrote to ChatGPT about his intentions to commit suicide.

According to the company’s website, the upcoming parental control features will allow parents to monitor how their children use ChatGPT. In addition, users will soon be able to add a trusted contact for emergency communication and contact emergency services with a single click.

OpenAI emphasised that ChatGPT already includes algorithms designed to discourage users from engaging in dangerous activities, but acknowledged that during prolonged interactions, the chatbot’s responses can sometimes be inaccurate.

The lawsuit, filed on August 26 in San Francisco state court by the parents of Adam Raine, alleges that OpenAI and CEO Sam Altman knowingly prioritised profit over user safety when they launched the GPT-4.0 version last year.

According to the complaint, Adam Raine died on April 11 after months of discussions with ChatGPT about suicide. The Raines allege that the chatbot validated Adam’s suicidal thoughts, provided detailed instructions for lethal self-harm, advised him on hiding alcohol and concealing a failed suicide attempt, and even offered to draft a suicide note.

The lawsuit seeks to hold OpenAI accountable for wrongful death and violations of product safety laws, while requesting unspecified monetary damages.

An OpenAI spokesperson expressed sorrow over Raine’s death, noting that the chatbot includes safeguards such as directing users to crisis helplines. The spokesperson added, “While these safeguards work best in short, common exchanges, we’ve learned that in extended interactions, some parts of the model’s safety training may degrade.” OpenAI plans to continually improve these protections.

According to the lawsuit, OpenAI launched GPT-4o in May 2024 with features that remembered past interactions, mimicked human empathy, and validated user emotions—capabilities that, without proper safeguards, put vulnerable users at risk. The Raines claim that this launch simultaneously increased OpenAI’s valuation from $86 billion to $300 billion and contributed to Adam’s death.

The Raines are seeking an order requiring OpenAI to verify users’ ages, refuse inquiries about self-harm methods, and warn users about the risk of psychological dependency on the chatbot.

By Tamilla Hasanova