OpenAI chief executive Sam Altman has announced that the company is working on enhanced protections for teenagers using ChatGPT, as U.S. lawmakers intensify scrutiny of the mental health risks posed by AI companions.
In a blog post published Tuesday, Altman said OpenAI is developing an age-prediction system to distinguish under-18 users from adults based on usage patterns, with a “safety-first” approach when there is uncertainty. “If there is doubt, we’ll play it safe and default to the under-18 experience,” Altman wrote, noting that in some countries the company may also request ID verification.
The company plans to restrict ChatGPT’s behavior with teenage users, including blocking flirtatious exchanges and preventing discussions of suicide or self-harm, even in fictional or creative contexts. Altman emphasized that if a teen user expresses suicidal thoughts, OpenAI will attempt to notify parents and, if necessary, contact authorities in cases of imminent risk.
These measures build on OpenAI’s recently announced parental controls, which include account linking with parents, disabling chat history, and alert systems that notify guardians if ChatGPT detects a user in distress.
The renewed focus on youth safety follows a lawsuit filed by the family of Adam Raine, a teenager who died by suicide after prolonged interactions with ChatGPT. During Tuesday’s Senate hearing, Adam’s father, Matthew Raine, delivered an emotional testimony, saying:
“ChatGPT spent months coaching him toward suicide. As parents, you cannot imagine what it’s like to read a conversation with a chatbot that groomed your child to take his own life.”
Raine claimed that suicide was referenced 1,275 times in his son’s conversations with the chatbot and urged Altman to suspend GPT-4o until OpenAI can guarantee its safety.
Another parent, speaking under the pseudonym Jane Doe, warned lawmakers that the issue has escalated into “a public health crisis,” calling it “a mental health war” that society is currently losing.
Altman acknowledged the emotional weight of these testimonies and reiterated OpenAI’s commitment to strengthening its safeguards to ensure its technology does not harm vulnerable users.
Recent Random Post:
