Meta is expanding its use of AI to find underage users on Instagram and will begin overriding account settings to protect them. AI for age detection has been used on the platform since 2024, but now the AI will proactively look for teen accounts with an adult birthday and get their settings right. The feature begins testing in the US on April 21.
Teenagers’ accounts on Instagram are subject to stricter rules by default. By default, teen accounts are private, can’t be reached by strangers, and are protected by Instagram about what they can see on the platform. So far, the systems have been looking at things like engagement data and even messages from friends, but that hasn’t curbed the issue.
Going forward, if an account is deemed to be a teenager’s, but the birthday registered on the platform indicates the owner is an adult, Instagram will automatically place it under teen restrictions — affected users will have the option to revert Instagram’s changes in case the AI makes a mistake.
Instagram isn’t relying on its AI systems alone to get teens to stop lying about their age online. Starting with the AI tech test, parents will also start receiving notifications on Instagram with “information about how they can have conversations with their teens on the importance of providing the correct age online.”
These changes and restrictions come as lawmakers and parents around the world are calling for better online security measures to protect their children. Since the teen accounts feature launched, Instagram has enrolled at least 54 million teenagers globally, with 90 percent of teens aged 13-15 “electing to remain in these protections.” The feature has also been extended to Facebook and Messenger.
There have been investigations concluding that Instagram recommends explicit content to kids on its platform. The EU also launched an investigation to check if Meta was adequately protecting children on its platforms, and the social media giant has gone through its fair share of trials following CSAM and pedophilic reports.
In the News: Google OAuth flaw exploited to bypass Gmail security filters
