Meta Disables AI Feature After It Harvested Family Data
Parenting blogger exposes AI scraping children's details from old posts.
1The Viral Exposure
Parenting blogger Kalie Robins revealed on September 2 how Meta's AI generated invasive prompts like 'Who's the child passenger?' under her Instagram video. The AI compiled details about her children by accessing multiple past posts, including deleted content. Robins' video garnered 308,000+ likes, sparking widespread outrage about AI privacy risks for minors.
2How the AI Worked
Meta's AI feature automatically suggested personal questions by scanning users' historical posts and connected profiles. In Robins' case, it extracted location data and even referenced her mother's account. While Meta claims the tool only uses accessible data, it demonstrated how AI can reconstruct private lives from scattered digital breadcrumbs within seconds.
3Meta's Damage Control
After CNET's inquiry, Meta stated it fixed the issue preventing AI from suggesting personal questions. The company admitted the feature 'missed the mark' but maintained it was designed to help users find information. However, Robins disputed Meta's claim about accessing a years-deleted photo, revealing deeper concerns about data retention policies.
4Unanswered Questions
In a September 10 follow-up, Robins demanded specifics: Which safeguards failed? What protections now exist for children's data? Meta hasn't directly apologized or explained how its AI compiled minors' information across accounts. The incident reveals transparency gaps in how generative AI systems process sensitive personal data without explicit consent.
5Ongoing Functionality Risks
Even after disabling personal prompts, Meta AI still suggests content based on publicly available data including web images and users' social media histories. This leaves open possibilities for unintended data recombination, especially concerning children whose digital footprints are often created by parents without considering long-term AI implications.
6Broader Privacy Backlash
The incident amplified existing criticism of Meta's handling of children's data, following 2024's multibillion-dollar settlements over youth addiction and exploitation risks. It also coincides with scrutiny of Meta's smart glasses' facial recognition tech, creating a pattern of privacy concerns around the company's AI ambitions.
7Parental Alarm Bells
Business Insider reports parents are now reconsidering sharing children's photos online after seeing how AI can repurpose such data. Cybersecurity expert Cathy Pedrayes called the incident 'creepy,' highlighting growing awareness about how seemingly innocent posts become training data for invasive AI behaviors with unpredictable consequences.
8The Bigger Picture
This case exemplifies the tension between generative AI's capabilities and personal privacy rights. As Meta rolls out Muse AI agents promising customized experiences, it faces mounting responsibility to implement robust safeguards—especially for vulnerable groups. The incident may accelerate calls for stricter regulations on AI data practices affecting minors.
Original article: CNET — Meta Disables AI Feature After It Harvested Family Data