Facebook and Instagram have a new way to detect and remove users under 13: AI bone structure analysis. In a blog post on Tuesday, Meta — Facebook and Instagram’s parent company — says its AI system will scan photos and videos posted to its platforms for “general themes and visual cues,” including height and bone structure.

“We want to be clear: this is not facial recognition,” Meta says in the blog post, adding that it “does not identify the specific person in the image.” This system is part of Meta’s efforts to keep kids under 13 off its platforms, and will also analyze posts, comments, bios, and captions to search for “contextual clues” that someone might be underage.

Meta’s AI-powered facial analysis, which is only available in “select” countries including the US ahead of a wider rollout, seems similar to the face-scanning tech offered by age verification services like Yoti and k-ID. Facebook and Instagram will deactivate accounts identified as underage, and the owner will need to verify their age to prevent it from deletion.

The announcement comes just days after a New Mexico jury found that Meta violated state law by misleading customers about the safety of its platforms and failing to protect children from child predators. Meta must pay $375 million as a result, and may have to implement changes that the company has already threatened to leave the state over.

Separately, Meta is expanding the technology it uses on Instagram to automatically identify and place users between 13 and 17 into Teen Accounts. These accounts come with stricter content controls, block messages from strangers, and prevent users under 16 from livestreaming. Instagram rolled out the tech first, and now Facebook will do the same for users in the US, followed by a rollout in the UK and EU in June.

In its announcement, Meta continues to advocate for age verification at the app store and operating system level, an approach that’s gaining traction in Congress and some states, including California and Colorado.