Meta has switched on its AI age estimation for Instagram worldwide. From 9 November 2026, the system looks for accounts it thinks belong to teenagers, even if they signed up with an adult birthday, and moves them into Teen Account protections.
For most adults nothing changes. But if the AI gets your age wrong, your account can suddenly go private, lose messaging options or hit a 60-minute reminder.
This guide explains what Meta announced, how the age AI works, what changes for flagged accounts, how to fix a wrong age with a Yoti video selfie or ID, what parents should do, and how Instagram compares with YouTube and Discord.
- What: Meta’s AI age estimation now runs on Instagram globally (announced 9 Nov 2026). Facebook is still rolling out, and Threads gets it “later this year”.
- How it works: AI reads profile signals such as birthday posts, school-grade mentions and activity, and in some countries analyses photos for general visual age cues. Meta says it is not facial recognition.
- If flagged as a teen: the account is moved into Teen Account settings: private by default, strict messaging, sensitive-content limits, a 60-minute reminder and sleep mode from 10 pm to 7 am.
- If flagged as under 13: the account is deactivated until you prove your age, or it is deleted.
- Wrongly flagged adult? Confirm your age through Meta’s partner Yoti with a video selfie, or upload a government ID. Yoti deletes the data after the check.
- Parents (US): an in-feed explainer video starts this month, linking to ConnectSafely guidance.
- Australia: Meta also uses Apple’s age-range signal from parent-set-up iPhones.
What Meta Announced on 9 November
In a Meta Newsroom post titled “Expanding Teen Account Protections to Teens Around the World”, the company said it is expanding its AI age-estimation technology globally on Instagram.
Meta’s goal, in its own words, is that “millions more teens across the world who use Instagram will receive these built-in protections,” including teens who entered an adult birthday.
| Platform | Status (9 Nov 2026) |
|---|---|
| AI age estimation expanding to every country | |
| Rollout continuing (US since May; UK and EU from June 2026) | |
| Threads | Similar technology “later this year” |
| Messenger | Teen Accounts exist; AI placement not separately announced |
Meta also said US parents will see an in-feed video on Instagram this month explaining the system. It links to an article by Larry Magid, CEO of the safety charity ConnectSafely, which is available to parents worldwide.
It lands in a busy month for Meta’s AI work, which also brought Muse Spaces mini apps to the Meta AI app.
How We Got Here: Timeline
| Date | What happened |
|---|---|
| Sep 2024 | Instagram launches Teen Accounts |
| Apr 2025 | Meta starts testing proactive AI placement of suspected teens in the US |
| Jun 2025 | Teen Accounts available in every country where Instagram runs |
| Sep 2025 | AI placement testing expands to the UK, Canada and Australia |
| 29 Apr 2026 | European Commission preliminary finding that Instagram and Facebook breached the Digital Services Act on under-13s |
| 5 May 2026 | AI placement on Instagram expands to all 27 EU countries and Brazil; visual analysis added in select countries |
| Aug 2026 | Meta agreement with 52 US attorneys general to invest in better teen and under-13 detection |
| 9 Nov 2026 | Global Instagram expansion announced |
How Instagram’s Age AI Works
Meta’s November post does not list the signals. Its May 2026 announcement and help pages describe several layers that work together.
- Profile and text signals: AI scans posts, comments, bios and captions for clues like “happy 15th birthday” or a school grade. Meta is extending this to Reels, Live and Facebook Groups.
- Visual analysis (select countries): AI looks at photos and videos for general cues, which press reports describe as height and bone structure, to estimate a general age.
- Activity patterns: how an account behaves on the app.
- Repeat-offender checks: new accounts made by people previously flagged as underage.
- Reports: user reports of underage accounts, many now reviewed by AI models.
Note: Meta says “this is not facial recognition.” A July 2026 clarification adds that the AI “does not use biometric data to identify whether a person is underage” and “does not create facial fingerprints.”
The output is a judgement that an account is likely a teen or likely under 13. Meta has not published accuracy or false-positive rates.
What Changes When You Are Placed in a Teen Account
A flagged account gets the same built-in protections as any Teen Account. Based on Meta’s published settings, that means:
| Setting | Teen Account default |
|---|---|
| Profile | Private |
| Messages | Strictest setting; only people they follow or are connected to |
| Sensitive content | Strictest control, 13+ content setting |
| Tags and mentions | Only from people they follow |
| Hidden words | Offensive terms filtered in comments and DM requests |
| Time limit reminder | Prompt to leave after 60 minutes a day |
| Sleep mode | Notifications muted 10 pm to 7 am |
Teens under 16 need a parent’s permission to loosen these settings, approved through Instagram’s supervision tools. Meta’s published rules do not require that approval for 16 and 17 year olds.
Tip: if you are an adult and your profile suddenly turns private or your DMs from strangers stop, check for a Teen Account notice in the app before changing anything else.
How to Fix It if Instagram Thinks You Are a Teen
Meta says people placed in the wrong experience can correct it. Changing an age from under 18 to over 18 requires proof through an ID or Yoti’s facial age estimation. Menu names vary a little by app version, but the flow is:
- Open the notice. If Instagram shows a “We think you may be under 18” or Teen Account message, tap it. It usually leads straight to age confirmation.
- Or go to your birthday setting. Profile, then the menu, then Accounts Center, then Personal details, then Birthday. Enter your real date of birth.
- Choose how to confirm. Pick a video selfie or upload ID when asked.
- Video selfie (Yoti): follow the on-screen prompts to record a short video of your face from a few angles. Yoti’s AI estimates your age and sends Meta only the result.
- ID upload: photograph a government ID such as a passport, driving licence or national ID card. Meta reviews it, which historically took up to about two days.
- Wait for the result. Once confirmed, the Teen Account protections are removed and you can set your own privacy and messaging options again.
Note: Yoti’s estimate has a margin of error of roughly 1.5 years for ages 13-19, according to earlier coverage of Instagram’s tests. If you are 18 or 19 and the selfie fails, the ID route is more reliable.
Where are the Yoti and ID options?
Australia’s eSafety Commissioner lists the same process for Instagram, Facebook and Threads: wrongly flagged users aged 16+ confirm their age through Yoti with a video selfie or a government ID. Yoti deletes the data once the check is done.
If Your Account Is Disabled as Under 13
This is the more serious case. Meta says accounts it believes belong to under-13s are deactivated, and the holder must prove their age through Meta’s verification process to stop the account being deleted.
- Act quickly: follow the link in the deactivation notice or email from Instagram.
- Use the same options: video selfie through Yoti or a government ID.
- No separate appeal form: Meta’s published routes are the age check itself, not a written appeal.
- Keep evidence: note the date of the notice and screenshot any error messages in case you need to contact support.
Warning: only use links inside the Instagram app or emails from Meta’s official domains. Scammers copy “verify your age” messages to steal logins and ID photos.
What Parents Should Do
Meta frames this as a backstop, not a replacement for parents. If you have a teen on Instagram:
- Check their birthday. Make sure the age on the account is correct, so protections apply without guesswork.
- Turn on supervision. Use Meta’s Family Center to link accounts, see time spent and approve setting changes for under-16s.
- Watch for the US explainer video. It appears in-feed this month and links to ConnectSafely’s guide on Meta’s “multi-layered age assurance approach”.
- Use device-level age settings. On iPhone, setting up the device as a child account lets Apple share an age range with apps that use it (currently Meta in Australia).
- Talk about it. Explain why the settings exist, so a teen does not try to bypass them with a second account, which Meta also checks for.
Worried about Meta’s AI more generally? Our guide on how to turn off Meta AI on WhatsApp, Instagram and Facebook covers the AI chat features and training opt-outs, which are separate from age checks.
Apple’s Age-Range Signal in Australia
Meta says it has started using Apple’s age-range signal in Australia. When a parent sets up a teen’s iPhone, they can have the App Store confirm the user’s age and assign an age range.
Each time the teen downloads an app, Apple passes that range to the developer, which can confirm whether the user meets the app’s age rules. Meta says it uses the signal voluntarily and that its own AI acts as a backstop.
Meta wants laws that make app stores verify age and get parental consent once per device. It argues this is “far more effective and privacy-preserving than asking every platform to verify age individually,” and it has publicly urged TikTok and YouTube to follow its approach.
Instagram vs YouTube vs Discord Age AI
Instagram is not alone. Several big platforms now guess your age with AI and ask you to prove it if they get it wrong.
| Instagram (Meta) | YouTube (Google) | Discord | |
|---|---|---|---|
| Signals | Profile text, activity, visual cues in select countries, reports | Searches, watch categories, account age | Social graph, account metadata, engagement counts, device data; no message content |
| Reads your face? | Visual analysis in some countries (Meta says not facial recognition) | No, only if you choose a selfie check | No, only if you choose facial estimation |
| Where | Instagram worldwide (Nov 2026) | US, UK, Australia, Brazil, Singapore, Switzerland, EEA | Selected regions with legal rules |
| If flagged | Teen Account; under-13 deactivated | Teen experience (no personalised ads, wellbeing limits) | Teen experience, age-restricted spaces blocked |
| Fix options | Yoti video selfie or government ID | Government ID, selfie or credit card | Facial age estimation or ID plus selfie |
| Data kept? | Yoti deletes after check | ID image removed, birthdate kept | Not specified in Discord’s explainer |
Sources: Meta Newsroom and eSafety for Instagram; All About Cookies (citing Google’s help pages) for YouTube; Discord’s safety explainer (22 Sep 2026) for Discord.
Privacy: What Meta Says It Does and Does Not Do
- Does: analyse what is already on your profile, how you use the app and, in some countries, general visual age cues in photos and videos.
- Says it does not: identify who is in an image, create facial fingerprints or use biometric data to decide if someone is underage.
- Verification data: Yoti handles video selfies and deletes them after the check, per Meta’s information to Australia’s regulator.
- Automated decisions: many underage reports are now reviewed by AI rather than a person, which Meta says is more accurate and faster in its testing.
If you want to check whether a profile photo or viral image is AI-generated rather than real, our guide on how to tell if a picture is AI walks through SynthID and Content Credentials.
What We Still Do Not Know
- Accuracy: Meta has not published how often the AI wrongly flags adults or misses teens.
- Which countries get visual analysis: Meta only says “select countries”.
- Exact signals and thresholds: not disclosed, to stop people gaming the system.
- Threads date: only “later this year”.
- EU outcome: the Commission’s April 2026 findings could lead to fines of up to 6% of Meta’s global turnover if confirmed.
My Take
Using AI to catch teens who lie about their age is a sensible backstop, and Meta’s layered approach is less invasive than forcing every user to upload an ID. For most adults, it will be invisible.
The weak spot is transparency. Without published error rates or a written appeal route, wrongly flagged adults, especially young-looking 18-20 year olds, carry the burden of proving their age with a selfie or ID.
My advice: make sure your birthday is correct now, and use only the in-app flow if you are ever asked to verify.
FAQ
Why did my Instagram account suddenly go private?
Meta’s AI may have placed you in a Teen Account. Look for a notice in the app and confirm your age with a Yoti video selfie or ID if you are an adult.
Does Instagram scan my face?
In select countries, AI analyses photos and videos for general visual age cues. Meta says it is not facial recognition and does not identify who is in the image.
How do I prove I am over 18 on Instagram?
Change your birthday in Accounts Center if needed, then confirm with a Yoti video selfie or a government ID when prompted.
Does Yoti keep my selfie?
Meta told Australia’s eSafety Commissioner that Yoti deletes the selfie or ID information once the check is complete.
What happens to accounts flagged as under 13?
They are deactivated. The holder must prove their age through Meta’s verification process, or the account is deleted.
Is this happening on Facebook and Threads too?
Facebook’s rollout is continuing, and Threads gets similar technology later in 2026, according to Meta.
Can teens turn the protections off?
Under-16s need a parent’s approval through supervision tools to loosen the settings. Older teens have more control, but the defaults still start strict.
Sources
- Meta Newsroom: “Expanding Teen Account Protections to Teens Around the World” (9 Nov 2026), about.fb.com/news/2026/11/expanding-teen-account-protections-to-teens-around-the-world/
- Meta Newsroom: “New AI-Powered Age Assurance Measures to Place Teens in Age-Appropriate Experiences” (5 May 2026, updated 31 Jul 2026)
- eSafety Commissioner (Australia): Information from age-restricted platforms (Meta’s Yoti video selfie and ID process)
- Unite.AI: “Meta Takes AI Age Estimation Global for Instagram Teen Protections” (9 Nov 2026): timeline and Teen Account settings
- Anadolu Agency: “Meta announces global expansion of AI age checks to protect teens on Instagram” (9 Nov 2026)
- PPC Land and The Decoder: coverage of Meta’s May 2026 visual analysis and EU/Brazil expansion
- Android Police: Instagram age verification options and Yoti accuracy (2022 test coverage)
- All About Cookies: YouTube AI age estimation and verification methods (Aug-Sep 2026)
- Discord: “How Discord Estimates Age Without Collecting Your Identity” (22 Sep 2026)
