ChatGPT Introduces Age Prediction Feature to Protect Minors from Sensitive Content

OpenAI is introducing a significant new feature in ChatGPT aimed at predicting the age of users on consumer plans. This enhancement is specifically designed to identify accounts that are likely owned by individuals under the age of 18, ensuring that the platform can implement age-appropriate experiences and necessary safeguards.

This update builds upon existing measures for users who self-identify as under 18 during the registration process. Such users are automatically provided with additional protections, which aim to minimize exposure to sensitive or harmful content. Meanwhile, adult users continue to enjoy the complete ChatGPT experience within the established safety guidelines.

Understanding ChatGPT’s Age Prediction Mechanism

The mechanism employed for age prediction by ChatGPT leverages a sophisticated model that assesses various behavioral and account-related signals. This includes factors such as the age of the account, typical usage patterns throughout the day, and the user’s self-reported age.

OpenAI states that by utilizing this model, they will gain insights into which factors contribute to greater accuracy, allowing continuous improvements to the system.

ChatGPT age prediction — ChatGPT Age Prediction Feature

In cases where users are mistakenly categorized under the under-18 experience, they can verify their age through a selfie-based verification process using Persona, a secure identity verification service. Additionally, users can check the status of safeguards on their account by navigating to the Settings and then Account section.

When an account is flagged as likely belonging to a minor, ChatGPT automatically transitions that user into a more secure experience. This transition limits access to various categories of content that are deemed higher risk for younger users, including:

Graphic violence or disturbing material
Potentially harmful viral challenges
Sexual or violent role-playing scenarios
Content depicting self-harm
Messages reinforcing extreme beauty standards or body shaming

The implementation of this feature is guided by expert recommendations and academic research on child development, which highlights differences in risk perception, impulse control, peer dynamics, and emotional regulation in adolescents. In instances where age-related information is unclear, ChatGPT will default to a safer interactive environment.

Beyond these automated safeguards, parents have the option to further tailor their teen’s experience using parental control tools. These features allow for the establishment of quiet hours, management of functionalities like memory and model training, and notifications if any alarming signs of distress are detected.

This rollout comes on the heels of OpenAI’s recent unveiling of ChatGPT Health, another new feature aimed at aiding users in understanding their health conditions.

Source & Images