Mark Zuckerberg’s Meta is implementing AI technology to identify underage Instagram users who misrepresent their age, automatically transitioning them to more protective account settings as part of its enhanced youth safety initiative.
Bloomberg reports that Meta is claiming to protect young users through the introduction of an “adult classifier” AI tool, designed to analyze user behavior and identify individuals under 18 years old regardless of their stated age. According to Allison Hartnett, Meta’s director of product management for youth and social impact, this proprietary software will examine various data points including user profiles, follower lists, content interactions, and even birthday-related posts to determine a user’s actual age.
The technology will automatically place suspected underage users into teen accounts, which feature enhanced privacy protections introduced by Instagram in September. These accounts implement stricter default settings, particularly regarding messaging capabilities and content accessibility. Users aged 16 or 17 will retain the ability to modify these restrictive settings, while those under 16 will require parental consent for any modifications.
The implementation of this technology comes at a crucial time for Meta, which faces significant legal challenges regarding its impact on youth mental health. The company is currently defending against a lawsuit from multiple state attorneys general alleging its role in fostering social media addiction among young users and contributing to a teen mental health crisis. These concerns gained prominence following whistleblower Frances Haugen’s 2021 disclosure of internal research suggesting Instagram’s negative effects on teenage girls’ mental well-being.
Meta’s age verification system will employ multiple mechanisms to prevent circumvention. The platform will flag attempts to create new accounts using the same email address with different birthdates and can track device IDs to identify account creators. Users attempting to manually increase their age will need to provide formal identification or submit a video-selfie to Yoti, a third-party service that estimates age based on facial features, with both companies deleting verification images after processing.
Meta’s approach leverages its extensive user data, including social connections, interests, and content preferences. While the company hasn’t disclosed the accuracy rate of its adult classifier, it has indicated that users incorrectly identified as teens will eventually have access to an appeals process, though this system is still under development.
Read more at Bloomberg here.
Lucas Nolan is a reporter for Breitbart News covering issues of free speech and online censorship.
