Buried in Meta's $18B settlement is a legal pass on kids' data
TechCrunch
β’Thu, 27 Aug 2026 20:04:26 +0000
π° What Happened
Buried inside Meta's settlement with attorneys general from 29 US states, which pays out up to 18 billion dollars, is a provision granting the states a legal pass for Meta's retention and use of children's data. The states agreed not to sue Meta under existing child safety laws over that data use, in exchange for guardrails tied to building an age-assurance model. Meta must develop, train and begin testing a system to detect which users on its platforms are under 13 within one year. The provision shields even that model-building data use from COPPA claims, a notable decision in a case framed around child safety.
π The Backstory
The settlement resolves coordinated privacy actions against Meta over the ways its platforms collect and handle children's information. COPPA (the Children's Online Privacy Protection Act) normally restricts collection and retention of children's personal data, which can collide with training a model that learns to detect minors. To bridge the gap, the states approved a permanent release of past, present and future COPPA claims tied to that specific use, while barring the use of kids' data for ad targeting or marketing. The result is that Meta gets a formal allowance to keep age-detection data regulators would otherwise expect it to delete.
π― Why It Matters
An agreement centered on child safety that quietly grants an exemption from child-safety law has drawn sharp criticism from privacy advocates. It sets a precedent for how settlement agreements could legitimize data use needed to train AI systems, even when the data involves minors. It raises hard questions about enforcement, since the guardrails may prove difficult to monitor in practice. For regulators, parents and competitors, it frames the trade-offs between effective age-assurance technology and the privacy rules intended to protect young users.
Buried inside Meta's settlement with attorneys general from 29 US states, which pays out up to 18 billion dollars, is a provision granting the states a legal pass for Meta's retention and use of children's data. The states agreed not to sue Meta under existing child safety laws over that data use, in exchange for guardrails tied to building an age-assurance model. Meta must develop, train and begin testing a system to detect which users on its platforms are under 13 within one year. The provision shields even that model-building data use from COPPA claims, a notable decision in a case framed around child safety.
The settlement resolves coordinated privacy actions against Meta over the ways its platforms collect and handle children's information. COPPA (the Children's Online Privacy Protection Act) normally restricts collection and retention of children's personal data, which can collide with training a model that learns to detect minors. To bridge the gap, the states approved a permanent release of past, present and future COPPA claims tied to that specific use, while barring the use of kids' data for ad targeting or marketing. The result is that Meta gets a formal allowance to keep age-detection data regulators would otherwise expect it to delete.
An agreement centered on child safety that quietly grants an exemption from child-safety law has drawn sharp criticism from privacy advocates. It sets a precedent for how settlement agreements could legitimize data use needed to train AI systems, even when the data involves minors. It raises hard questions about enforcement, since the guardrails may prove difficult to monitor in practice. For regulators, parents and competitors, it frames the trade-offs between effective age-assurance technology and the privacy rules intended to protect young users.