OpenAI introduces parental controls for ChatGPT following California teen suicide

New Delhi: OpenAI is launching parental controls for its ChatGPT platform on both web and mobile starting Monday, in response to a lawsuit filed by the parents of a California teen who died by suicide. The lawsuit claims the AI chatbot had allegedly provided guidance on methods of self-harm.

The new parental controls allow parents and teenagers to opt in for enhanced safeguards by linking their accounts. The controls activate only after mutual consent between the teen and parent, according to OpenAI.

U.S. regulators have been increasingly focusing on potential negative impacts of AI chatbots. In August, reports highlighted how Meta’s AI systems permitted inappropriate flirty conversations with minors.

Under OpenAI’s updated measures, parents can limit their teen’s exposure to sensitive content, manage whether ChatGPT retains memory of previous conversations, and decide if interactions are used to train OpenAI’s models. Parents can also establish quiet hours, disable voice mode, and block image generation or editing. However, they will not have direct access to their teen’s chat transcripts.

In rare cases where the system or trained reviewers detect a serious safety concern, parents may receive minimal information necessary to ensure the teen’s safety, and they will be notified if the teen unlinks accounts.

With around 700 million weekly active users of ChatGPT, OpenAI is also developing an age-prediction system to identify users under 18, allowing the chatbot to automatically apply teen-appropriate settings.

Meta has similarly rolled out additional safeguards for its AI products last month, aiming to prevent flirty conversations and discussions about self-harm or suicide with minors, while temporarily restricting access to certain AI characters.