MENLO PARK, California — Meta Platforms said on Monday 9 November 2026 that it will expand its artificial intelligence age-estimation technology around the world to move millions more teenagers on Instagram into Teen Account protections, even when the birth dates on their accounts say they are adults, as the company faces a wave of lawsuits alleging that its platforms harm young users.
The announcement, reported by Reuters on Monday, closes one of the oldest loopholes in social media safety: the birth date typed at sign-up. A teenager who entered an adult birthday could slip past the stricter settings reserved for 13- to 17-year-olds. Under the expanded system, Meta's software will weigh the evidence of an account's real age and move suspect accounts into Teen Accounts.
Illustration: A teenager holding a smartphone showing the Instagram app with a shield and checkmark, beneath a glowing AI scanning overlay. The image is an AI-generated illustration, not a photograph of a real person or event.
What Happened
Meta said on Monday 9 November 2026 that it would expand its AI age-estimation technology globally on Instagram, placing millions more teenagers into Teen Account protections. The company described the system as technology "to find people we suspect to be teens, even if they use an adult birthday," according to its announcement. An account judged likely to belong to a teenager will now be switched into the stricter settings automatically.
Teen Accounts, available on Instagram, Facebook and Messenger, automatically limit who can contact teenagers, the content they see, and the time they spend on the apps, according to Reuters. Accounts are private by default, messaging is restricted to people the teenager follows or knows, and sensitive content is filtered at the strictest setting. Teenagers under 16 need a parent's permission to loosen any of these settings.
The system does not ask users for their age. Instead, it estimates it. The software looks at signals such as the profiles and content an account interacts with, birthday messages it receives, mentions of school grades, and general visual cues in photos and videos, including height and bone structure. Meta has stressed that this is not facial recognition. In its own words, published in May 2026: "We want to be clear: this is not facial recognition. Our AI looks at general themes and visual cues, for example height or bone structure, to estimate someone's general age; it does not identify the specific person in the image."
Accounts believed to belong to users under 13 are removed, and the holder must prove their age to get the account back. Accounts judged to belong to 13- to 17-year-olds are moved automatically into Teen Accounts. Accounts judged to belong to adults are left alone. Meta said it is continuing to roll out the technology on Facebook and will begin launching similar technology globally on Threads later this year.
US parents will begin receiving Instagram notifications this month explaining the process, Meta said. In Australia, Meta has begun using Apple's age-range signal as an additional age check. The system is not new: Meta first deployed it in the United States, extended it to the EU and Brazil during 2025, and widened it again in May 2026.
Background
The programme at the centre of the announcement dates back to September 2024, when Meta launched Teen Accounts on Instagram. Announced on 17 September 2024, it rebuilt its apps around teenagers rather than bolting parental controls onto an adult product. New teenage sign-ups were placed into Teen Accounts immediately, and existing teenage accounts in the US, UK, Canada and Australia were moved over within 60 days, with teenagers worldwide following from January 2025, and later to Facebook and Messenger.
By June 2025, Meta said 54 million teenagers were enrolled on Instagram, and 97 percent of those aged 13 to 15 had stayed within the strict default settings. The programme worked at scale — but only for teenagers the company could identify. Anyone who had signed up with a false adult birthday stayed outside it.
Illustration: A parent and a teenage child looking together at a smartphone showing privacy and parental control settings. The image is an AI-generated illustration, not a photograph of a real person or event.
The second piece of background is the legal pressure Meta now faces. In August 2026, the company struck settlements worth up to $18 billion with nearly every US state and territory to resolve lawsuits brought by attorneys general over the platforms' harm to teenagers. Announced on 26 August 2026, it covered a bipartisan group of 52 attorneys general. Meta said about 70 percent — roughly $12.7 billion — would be paid over ten years, with the rest released only if YouTube and TikTok reach similar settlements. A federal judge in Oakland approved it on 26 August 2026. Under the settlements, Meta agreed to improve how it identifies teenagers and children under 13, and to accept a two-hour daily limit and a midnight-to-6 a.m. block. Meta denied any wrongdoing; the settlement admits no liability.
The third piece is the courtroom. Meta settled with a teenage plaintiff last month in a Los Angeles trial over claims that Instagram and Facebook were designed to be addictive, causing mental health problems and an eating disorder. Jury selection began on Wednesday 28 October 2026 for two teenagers, S.J., 16, from Illinois, and K.D.B., 18, from Mississippi. Meta had settled with S.J. days before the trial, and settled with K.D.B. on Friday 30 October, two days into jury selection, on undisclosed terms. Snap then settled too: on Monday 2 November, two days before opening statements, it reached a confidential settlement with both teenagers, leaving no defendant in the case, so the trial will not go ahead, according to Reuters. The case was chosen from thousands of similar lawsuits as a test case. In the first such trial, in March 2026, a jury found Meta and YouTube negligent and awarded $6 million to a young woman who said their platforms had harmed her mental health.
Australia offers a preview of where this enforcement leads. Australia's world-first law requiring platforms to keep under-16s from holding accounts took effect on 10 December 2025. Meta said it removed access to more than 750,000 Facebook and Instagram accounts in Australia that it assessed as belonging to people under 16, covering December 2025 to June 2026. Australia's eSafety Commissioner opened formal investigations into the major platforms in March 2026.
Why It Matters
The expansion matters first because it attacks the weakest link in every age-based safety system: self-declaration. A birthday typed into a sign-up form has always been an honour system, and teenagers have always known it. By judging accounts on behaviour instead, Meta is moving from asking users how old they are to inferring it — a shift that could put millions of teenagers now using adult accounts under the protections designed for them. For parents, the practical change is simple: a child who lied about their age may now find themselves placed in one anyway.
It matters, second, because the technology is probabilistic, not certain — and Meta has not published accuracy figures for the expanded system, according to The Verge. Every estimate can be wrong in two directions. An adult wrongly classified as a teenager gets restrictions they did not need. A teenager the system misses stays outside the protections entirely. How often each mistake happens, and how easily a wrongly flagged user can appeal, will decide whether the system feels like safety or surveillance. Meta has said wrongly placed users can seek to verify their age, for example with an ID, but has not published full details of appeals in every country.
Third, the visual analysis at the heart of the system raises privacy questions that will not go away simply because Meta says it is not facial recognition. Scanning photos and videos of children and teenagers for clues about their age means a company's AI is drawing conclusions about minors' bodies from their pictures. Privacy advocates have criticised the approach, and European regulators are likely to examine it closely under the EU's Digital Services Act. Meta's answer is that the analysis estimates an age range and does not create facial fingerprints or identify individuals.
Finally, the move sets a pattern for the whole industry. Meta is the largest social media company in the world, and its settlement was explicitly designed to pull rivals along: a large share of the $18 billion depends on YouTube and TikTok adopting similar measures. If Meta's age estimation works at global scale, other platforms will face pressure to do the same; if it fails visibly, it will become the cautionary tale that slows everyone down.
Illustration: A modern courthouse with a large gavel and scales of justice beside a smartphone showing a social media feed. The image is an AI-generated illustration, not a photograph of a real place or event.
What Happens Next
The immediate next step is the rollout itself. The expanded age estimation is now going live on Instagram worldwide, and the test will be visible quickly: teenagers who signed up with adult birthdays should begin finding themselves in Teen Accounts. Meta is continuing to roll out the technology on Facebook and will begin launching similar technology globally on Threads later this year. Watch, too, for whether Meta publishes accuracy figures or an appeals process.
In the United States, parents will begin receiving Instagram notifications this month explaining how teenagers are placed in age-appropriate protections, and the company says it is investing in stronger technology to identify children under 13 as well. The August settlement also has a timetable: within a year, Meta must have an age-verification system that can be independently checked for accuracy, plus a separate system to catch users under 13. Those commitments will be measured against the new global rollout.
In the courts, the Los Angeles trial will not go ahead: Snap settled with both teenagers on 2 November, following Meta, TikTok and YouTube, and thousands of similar lawsuits remain pending against Meta and the other major platforms. In Australia, the eSafety Commissioner's investigations continue. The expansion is Meta's answer to all of these pressures at once.
Sponsored: explore today's featured sponsored offers
FAQ
What exactly is Meta expanding?
Meta is expanding its AI age-estimation technology globally on Instagram, so accounts it suspects belong to teenagers are moved into Teen Account protections even if the birth date says the user is an adult. The company said the change would place millions more teenagers under those protections. The rollout continues on Facebook, with Threads to follow later this year.
How does the AI estimate a user's age?
The system looks for signals that suggest a user may be under 18, rather than trusting the birth date typed at sign-up. Those signals include the profiles and content an account interacts with, birthday messages it receives, mentions of school grades, and general visual cues in photos and videos, including height and bone structure. Meta says this is not facial recognition: it estimates an age range without identifying the person. Accounts believed to belong to users under 13 are removed; those believed to belong to 13- to 17-year-olds are placed in Teen Accounts.
What protections do Teen Accounts give teenagers?
Teen Accounts automatically limit who can contact teenagers, the content they see and their time on the apps. Accounts are private by default, with restricted messaging and the strictest content filtering. Time-limit reminders and a night-time sleep mode limit use. Teenagers under 16 need a parent's permission to loosen any of these settings. Meta launched Teen Accounts in September 2024; by June 2025 it said 54 million teenagers were enrolled.
Why is Meta doing this now?
The expansion comes as Meta faces intense legal pressure over harm to teenagers. In August 2026 it agreed to settlements worth up to $18 billion with 52 US attorneys general, promising to improve how it identifies teenagers and children under 13. In October 2026 it settled with two teenage plaintiffs in a Los Angeles trial over claims its platforms were designed to be addictive, and in March 2026 a Los Angeles jury found Meta and YouTube negligent in a similar case. Australia's under-16 law, in effect since 10 December 2025, has added to the pressure.
What happens if the AI gets my age wrong?
That is the open question. Meta has not published accuracy figures for the expanded system, according to The Verge, so it is not known how often adults are wrongly placed or teenagers missed. Meta has said wrongly placed users can seek to verify their age, for example with an ID, but has not published full details of appeals in every country. US parents will begin receiving Instagram notifications this month explaining the process.
Sources
- Reuters, 9 November 2026 — Meta expands AI age checks for Instagram teens as lawsuits mount. https://www.reuters.com/legal/litigation/meta-expands-ai-age-checks-instagram-teens-lawsuits-mount-2026-11-09/
- The Verge, 9 November 2026 — Meta rolls out bone-structure AI age estimation globally to shunt underage users into Teen Accounts. https://www.theverge.com/tech/1029627/meta-instagram-ai-age-assurance-global-expansion
- Meta, 26 August 2026 — Our agreement with bipartisan attorneys general: calling on TikTok and YouTube to join us in supporting teens (settlement of up to about $18 billion, 52 attorneys general, judge's approval 26 August 2026). https://about.fb.com/news/2026/08/agreement-with-state-attorneys-general-supporting-teens/
- Reuters, 30 October 2026 — Meta settles with teen during trial testing claims social media caused depression, eating disorders (Los Angeles trial; S.J. and K.D.B.). https://www.thehindubusinessline.com/info-tech/social-media/meta-settles-with-teen-during-trial-testing-claims-social-media-caused-depression-eating-disorders/article71648169.ece
- Reuters, 2 November 2026 — Snap settles with both teenagers two days before opening statements, ending the Los Angeles trial (reported by Reuters via NBC News and Bloomberg Law).
- Meta, 5 May 2026 — AI age-assurance update: visual analysis added, age-prediction technology extended to Instagram in the EU and Brazil and Facebook in the US; "this is not facial recognition" explanation.
- Instagram, 17 September 2024 — Introducing Instagram Teen Accounts (launch; private by default, messaging and content limits, sleep mode, parental permission under 16). https://www.scoop.my/news/228049/instagram-launches-new-teen-accounts-with-built-in-protection-for-minors/
- Digwatch, 13 August 2026 — Meta blocks access to 750,000 under-16 accounts in Australia (removals December 2025 to June 2026 under Australia's under-16 law, in effect 10 December 2025). https://dig.watch/updates/meta-australia-social-media-ban-750000-accounts
Comments
Post a Comment