Meta extends AI age estimation worldwide, moving millions more Instagram teens into protections

Global rollout of age detection

Meta said on 9 November 2026 that it is expanding its AI age estimation technology worldwide, a step it says will move "millions more" teenagers on Instagram into Teen Account protections. The system looks for accounts that appear to belong to people aged 13 to 17, even when the owner entered an adult birthday, and then limits who can contact them, what content they see and how long they spend in the app. The Verge reports that Meta first launched the age-assurance updates in the United States, the European Union and Brazil in 2025. In Australia, the company has also started using the age-range signal that Apple provides in its app store as an extra detection method.

Meta described the approach in its announcement:

“We use AI technology to find people we suspect to be teens, even if they use an adult birthday, and place them in Teen Account protections.”

How the detection works

The Verge says the tools look at the other profiles and types of content an account interacts with, the birthday messages it receives, and "general themes and visual cues" from uploaded photos and videos, including height and bone structure. Accounts believed to belong to children under 13, the minimum age for Meta's social accounts, are removed from the platform. Teen Accounts are private by default and prevent strangers from sending them direct messages. Users aged 16 or younger cannot change those settings without a parent's or guardian's permission, and parents can supervise accounts to set time limits and monitor use.

Timeline of the age checks

The first wave of protections dates to 2025, when the age-assurance updates launched in the United States, the European Union and Brazil. Meta agreed to settlements in August 2026 with nearly all US states and territories, which had sued over the effects of its platforms on teenagers, and committed to improve how it identifies teens and children under 13. In October 2026, Meta also reached a settlement with a teenage plaintiff in a Los Angeles trial over claims that Instagram and Facebook were designed to create dependence and caused mental health problems and an eating disorder. The global expansion was announced on 9 November 2026.

Parents, Threads and Facebook

Parents in the United States will see an in-feed Instagram video explaining how age-appropriate protections work and how they can help confirm their child's age. 9to5Mac and Mediafax date the first parent notifications to November 2026, while The Verge says they will start in December 2026. In Australia, when a parent sets up a child's phone, the app store can confirm the user's age and assign an age range that Meta now uses as an additional check. Meta says it will keep extending the technology on Facebook and begin rolling out similar technology on Threads toward the end of 2026. The Verge notes that automatic age checks reached US Facebook accounts in May, and Meta has not said whether they will expand to other regions.

Settlement figures and the EU draft

The reports differ on the size of the August settlement. The Verge gives $17.1 billion, while Mediafax, relaying Reuters, gives up to $18 billion. Both versions say Meta agreed to strengthen its age assurance practices and impose heavier restrictions on teen users. Separately, Bloomberg News reported on a draft of the EU Digital Fairness Act that would require companies to let adult users switch off certain features. The same report says the draft also targets AI chatbot features, but the text it quotes is cut off before the full list, so the scope of the proposal remains unclear from the sources.