

OpenAI has detailed how its age prediction system in ChatGPT works to provide additional safety protections for younger users. The system uses an AI model to estimate whether a ChatGPT account is likely to belong to someone under the age of 18. The company says it considers behavioural and account-level signals, including the general topics discussed, typical usage times, how the account is used, usage patterns over time and the age of the account.
The age prediction system does not replace the date of birth provided when creating a ChatGPT account. Instead, it acts as an additional safety check. If ChatGPT identifies an account as belonging to someone under 18, the ChatGPT for Teens experience is automatically enabled, which places certain restrictions on features and interactions while allowing users to continue learning, creating and asking questions. Adults who are incorrectly classified can verify their age through Persona, a third-party verification service. Depending on the country, verification may involve a live selfie, a government-issued ID or both.



















Comments (0)
No comments yet
Be the first to comment!