Meta AI app usage notifications spark privacy concerns

Meta’s consumer AI application, Meta AI, has climbed to No. 5 on the U.S. App Store, rising from its previous position at No. 57, according to market intelligence provider Appfigures. However, this growth has been accompanied by user privacy concerns. Specifically, Meta started sending people Instagram notifications about which of their friends were using the Meta AI app, presumably to encourage downloads. This cross-platform notification system has drawn scrutiny from users who were not aware their activity on the AI application would be broadcast to their social networks.

The spike in downloads follows Meta’s release of its new Muse Spark AI model on Wednesday. While the Muse Spark model is new, the Meta AI app itself might be new to many, though it launched last April. In its first month and a half on the market, the app recorded 6.5 million downloads, according to Appfigures. This is a relatively small figure for Meta, which counts an estimated 42% of the world as daily users of at least one of its apps. Because the Meta AI app requires users to log in with a Meta account, it connects directly to their existing Facebook and Instagram profiles. This deep integration allows Meta to use activity from the AI app, alongside Facebook and Instagram data, to serve targeted advertisements, raising questions about how user data is leveraged across the company’s ecosystem.

These privacy concerns are compounded by the app’s history of data exposure. Meta previously removed a “Discover” feed from the Meta AI app after design issues led to accidental public sharing of private chat logs. The issue was highlighted by a16z partner Justine Moore, who noticed that the feed was populated with conversations from users who did not realize their chats were public. While some shared logs were benign—such as when a man with a Southern accent, a user of the Meta AI app, asked, “Hey, Meta, why do some farts stink more than other farts?”—others exposed highly sensitive data. In these cases, people share their personal home address, information about medical issues, and intimate concerns about their marriage. Although users had to manually publish these chats, the high rate of accidental sharing prompted Meta to remove the Discover feed entirely.

Why it matters

Meta’s deep app integration makes it difficult for users to track data sharing, creating risks where private chat logs or personal details can be exposed or used for targeted advertising without clear consent.