The company is combining time limits, parental controls and age estimation as pressure mounts over how artificial intelligence affects younger users.
OpenAI is preparing a separate “ChatGPT for Teens” mode designed for users under 18. The new format will include additional restrictions and safety tools for teenagers amid lawsuits and increased scrutiny of chatbots’ impact on children.
The company said the mode will combine protective measures that have been gradually introduced over the past year. In particular, ChatGPT will suggest that teenagers take a break after 90 minutes of using the service within a three-hour period.
Parents and underage users will be able to set “quiet hours,” during which access to the chatbot will be disabled. “Study hours” will also be introduced, automatically activating learning mode.
What restrictions will apply in ChatGPT for Teens
OpenAI is tightening the rules governing the chatbot’s interactions with minors. The system is intended to prevent situations in which ChatGPT expresses personal feelings toward a user. Separately, the service will warn teenagers about the risks if they attempt to upload private or sensitive images.
Users who indicate they are under 18 during registration will automatically be placed under protected settings. At the same time, the company will use artificial intelligence to assess users’ ages. If the system determines that a person may be a minor, the restrictions will apply even if a different date of birth is listed in their profile.
Why OpenAI is strengthening protections for minors
According to a Pew Research Center study released in December, nearly one-third of American teenagers use AI-powered chatbots every day. ChatGPT was the most popular service among survey participants: more than half of respondents said they had used it.
OpenAI faces criticism over young people’s potential access to inappropriate content, the risk of increased loneliness, and the negative impact of AI services on education. The company denies claims made by some families in lawsuits linking their children’s conversations with ChatGPT to suicides.
Last fall, OpenAI added parental controls that allow users to turn off memory and chat history. Parents can also receive notifications if the system detects signs that a child is in acute emotional distress. In the coming weeks, the company plans to launch additional alerts for parents related to eating disorders.
Lawsuits against OpenAI and Sam Altman
In April, seven families of victims of the Tumbler Ridge school shooting in British Columbia filed a lawsuit against OpenAI and CEO Sam Altman. The plaintiffs claim that ChatGPT may have been connected to the injuries or deaths of their children, as the shooter allegedly discussed gun violence with the chatbot for an extended period.
Altman apologized to the families. In response, OpenAI said it had already strengthened safety restrictions in the service.
In June, the state of Florida filed a lawsuit against OpenAI and Altman, accusing them of failing to adequately address the risks ChatGPT may pose to minors. The lawsuit alleges possible assistance to mass shooters, encouragement of suicide, the creation of dependency without adequate parental oversight, and users’ loss of critical-thinking skills.
OpenAI said it had implemented industry-leading safeguards and policies for safer use of the chatbot.
Last week, the Massachusetts district attorney charged a 17-year-old teenager with two counts of murder in connection with the deaths of his mother and brother. According to prosecutors, before the incident, the young man discussed theoretical ideas or fantasies about killing his family with ChatGPT. OpenAI did not provide an immediate comment on the case.
The launch of ChatGPT for Teens is expected to be one of OpenAI’s steps in response to growing demands for the protection of children and teenagers when using generative artificial intelligence.