As concern for AI’s impact on young people continues to mount, OpenAI has introduced an “age prediction” feature into ChatGPT. This feature is designed to help identify minors and place sensible content constraints on their conversations.
OpenAI has faced significant criticism in recent years regarding ChatGPT’s potential effects on children. A number of teen suicides have been linked to the chatbot. Like other AI companies, OpenAI has also been criticized for allowing ChatGPT to discuss sexual topics with young users. Last April, the company addressed a bug that allowed the chatbot to generate erotic content for users under the age of 18.
The company has been working on its underage user problem for some time. This new “age prediction” feature adds to protections already in place. The feature uses an AI algorithm that assesses user accounts for specific behavioral and account-level signals to identify young users.
These signals include factors like the user’s stated age, how long the account has existed, and the times of day the account is usually active. OpenAI already employs content filters designed to block discussions of sex, violence, and other potentially problematic topics for users under 18. If the age prediction mechanism identifies an account as belonging to a minor, those filters are automatically applied.
If a user is mistakenly identified as underage, there is a process to reestablish an adult account. They can submit a selfie to OpenAI’s ID verification partner, Persona.