Instagram’s AI Age Check Flags an Adult—but Still Failed on a 13-Year-Old Under New Australian Rules
Meta’s video-selfie system passed a moustachioed man but rejected a test by a teenager — casting doubt on accuracy as the under-16 social media ban looms
Meta’s new age-verification system correctly identified a moustachioed adult as over sixteen — but failed when a thirteen-year-old tried the same process, exposing flaws under Australia’s impending under-16 social media ban.
The test, conducted by journalists ahead of the law coming into force on December tenth 2025, used a so-called “video selfie”, in which users move their head to allow facial-age estimation.
The adult’s face—complete with facial hair—was cleared within minutes, but the 13-year-old, even after altering the date of birth to appear older, triggered a denial: the system requested a government-issued ID.
Meta’s chosen provider, the age-assurance firm, says its AI models perform most reliably on users over 21, while adolescents and minority-group faces often produce higher error rates.
That suggests the facial-recognition test may not offer consistent protection for younger users who misstate their age.
Meta also told regulators it may rely on other signals — such as account history or user behaviour — to judge age, raising concerns about transparency and privacy.
The Australian trial comes as the federal government enacts the world’s first nationwide ban on social-media use by children under 16. Platforms including Instagram, Facebook and others must now implement robust age checks or face steep penalties.
The law has drawn both praise for protecting minors and warnings about potential flaws in the verification systems.
Privacy advocates and civil-liberties experts warn that unreliable face-scanning tools risk arbitrary or discriminatory blocking of legitimate users — and that the absence of a successfully verified ID option for some might entrench access inequality.
As the ban’s effective date nears, Meta’s system will face its toughest test yet: removing hundreds of thousands of under-sixteen accounts while ensuring compliance without false positives.
For now, the mixed results of the trial raise serious questions about whether AI-based age verification can deliver on its promise — or whether it risks locking out vulnerable users because of flawed estimates.
The outcome may shape global discussions on how best to protect minors online while preserving fairness and rights.
The test, conducted by journalists ahead of the law coming into force on December tenth 2025, used a so-called “video selfie”, in which users move their head to allow facial-age estimation.
The adult’s face—complete with facial hair—was cleared within minutes, but the 13-year-old, even after altering the date of birth to appear older, triggered a denial: the system requested a government-issued ID.
Meta’s chosen provider, the age-assurance firm, says its AI models perform most reliably on users over 21, while adolescents and minority-group faces often produce higher error rates.
That suggests the facial-recognition test may not offer consistent protection for younger users who misstate their age.
Meta also told regulators it may rely on other signals — such as account history or user behaviour — to judge age, raising concerns about transparency and privacy.
The Australian trial comes as the federal government enacts the world’s first nationwide ban on social-media use by children under 16. Platforms including Instagram, Facebook and others must now implement robust age checks or face steep penalties.
The law has drawn both praise for protecting minors and warnings about potential flaws in the verification systems.
Privacy advocates and civil-liberties experts warn that unreliable face-scanning tools risk arbitrary or discriminatory blocking of legitimate users — and that the absence of a successfully verified ID option for some might entrench access inequality.
As the ban’s effective date nears, Meta’s system will face its toughest test yet: removing hundreds of thousands of under-sixteen accounts while ensuring compliance without false positives.
For now, the mixed results of the trial raise serious questions about whether AI-based age verification can deliver on its promise — or whether it risks locking out vulnerable users because of flawed estimates.
The outcome may shape global discussions on how best to protect minors online while preserving fairness and rights.
AI Disclaimer: An advanced artificial intelligence (AI) system generated the content of this page on its own. This innovative technology conducts extensive research from a variety of reliable sources, performs rigorous fact-checking and verification, cleans up and balances biased or manipulated content, and presents a minimal factual summary that is just enough yet essential for you to function as an informed and educated citizen. Please keep in mind, however, that this system is an evolving technology, and as a result, the article may contain accidental inaccuracies or errors. We urge you to help us improve our site by reporting any inaccuracies you find using the "Contact Us" link at the bottom of this page. Your helpful feedback helps us improve our system and deliver more precise content. When you find an article of interest here, please look for the full and extensive coverage of this topic in traditional news sources, as they are written by professional journalists that we try to support, not replace. We appreciate your understanding and assistance.
