The Instagram logo. Image: Instagram / Wikimedia Commons, Public domain (trademarked), cropped Instagram will now use AI to guess whether you are really a teenager, wherever you live, and move anyone it suspects is under 18 into its stricter Teen Accounts, even if they signed up with an adult birthday. Meta said on Monday it is “expanding this technology globally”, and that “millions more teens across the world who use Instagram will receive these built-in protections”.
What is new today #
Until now the system only ran on Instagram in some countries. Meta first switched it on in the US, the UK, Canada and Australia, then in May added the 27 EU countries and Brazil, saying it aimed to take it global on Instagram “throughout the year”. Today’s step does that: Meta’s post names no countries left out, and says that “no matter where you live” it will try to place teens in age-appropriate settings.
Facebook and Threads come next. Meta says it is “continuing to roll this technology out on Facebook”, where it started in the US in May with the UK and EU to follow, and that “later this year” it will begin launching similar technology globally on Threads.
Two smaller changes arrive too. In Australia, Instagram has started using Apple’s age-range signal, which an iPhone passes to apps when a parent has set up a teen’s device. And this month parents in the US will see an in-feed video on Instagram explaining the approach, linking to an article on age checks by Larry Magid, the internet safety advocate who runs ConnectSafely, which parents anywhere can read.
How Instagram’s AI guesses your age #
Meta’s May post explains the method. Its AI reads whole profiles for clues such as birthday celebrations or mentions of school grades, in posts, comments, bios and captions, and it has added visual analysis, which scans photos and videos for signs that the person in them is young. Meta insists this is not facial recognition. A correction it added to that post on July 31 says the AI “does not create facial fingerprints or identify who is in a photo”. In May, Meta said visual analysis was available only in “select countries”.
What happens next depends on the age the AI settles on. Accounts it thinks belong to 13 to 17 year olds are switched to Teen Account protections, which limit who can contact them, what they see and how long they spend in the app. Accounts it thinks belong to children under 13, Instagram’s minimum age, are deactivated and deleted unless the holder proves their age.
What to do if Instagram gets your age wrong #
Monday’s post does not describe a new appeal route, so the existing one applies. If someone placed in teen settings tries to change their birthday from under 18 to 18 or over, Meta says it requires them to verify their age, either with an ID or with Yoti’s facial age estimation, a video selfie checked by the outside company Yoti.
Meta has said that Meta and Yoti delete a video selfie once the age is confirmed, and that an uploaded ID is stored on its servers and deleted within 30 days. Anthropic uses the same company to let adults wrongly flagged on Claude prove their age.
Meta wants app stores to do the checking #
The global push follows Meta’s August agreement with 52 US attorneys general over teen safety, under which it committed to stronger ways of identifying teens and children under 13. The Electronic Frontier Foundation called that deal “a bad deal for all internet users”, arguing it “firmly embeds deeply flawed age estimation technology” into millions of people’s online lives.
Meta used Monday’s post to argue that the job belongs with Apple and Google. It said it supports laws requiring app stores to verify age and get parental consent, because “one verified signal from people’s devices, shared consistently with all apps, is far more effective and privacy-preserving” than every app checking separately. It also called on TikTok and YouTube to adopt the same practices. Others are moving the same way: New York’s new chatbot law requires age checks from January 1, 2027, and Apple will rate AI companion apps 18+ on its App Store from November 16.
Why it matters #
An AI model now decides which version of Instagram people get in every country, and anyone it wrongly marks as a teen has to show an ID or a face scan to get out. Meta says it is “encouraged by the results”, but its post gives no figure for how often the system gets an age wrong.
Sources: Meta, Expanding Teen Account Protections to Teens Around the World (November 9, 2026); Meta, New AI-Powered Age Assurance Measures (May 5, 2026, corrected July 31); Meta, Introducing New Ways to Verify Age on Instagram (June 2022); Electronic Frontier Foundation (September 1, 2026).