Meta has introduced a disclaimer to its AI app, advising users against sharing personal or sensitive information in the app’s public feed.

This warning appears when users choose to share their interactions publicly, responding to ongoing privacy criticisms.

The update follows reports of users unintentionally sharing private or embarrassing information in the app’s “discover” feed.

These posts sometimes included personally identifying details, raising concerns among privacy advocates.

Although the Meta AI app, launched in April, does not share chat histories by default, users can make their interactions public without fully understanding the risks.

Privacy experts have criticized the app’s design for lacking clear indicators about public sharing. Security expert Rachel Tobac pointed out that users typically do not expect AI chatbots to feature a public social feed.

With the new update, users are warned that shared prompts are public and may be visible across other Meta apps. However, this disclaimer only appears on the first public share.

Meta has not commented on whether further changes to the app are planned.

🔗 Source: Engadget

🧠 Food for thought

1️⃣ Meta’s reactive pattern to privacy issues persists despite years of regulatory action

The Meta AI sharing issue follows a well-established pattern for the company: launch features with privacy concerns, face public backlash, then implement fixes afterward.

This approach dates back to Facebook’s 2006 News Feed launch, which sparked protests from over 1 million users concerned about privacy, and continued with the controversial Beacon tracking system in 2007 that shared user purchases without consent 1.

Despite a $5 billion FTC penalty in 2019 (the largest ever for privacy violations) and requirements to implement new privacy structures and oversight mechanisms, Meta continues to launch products with problematic privacy designs 2.

The company has faced fines totaling €395 million from European regulators in 2022-2023 for illegal data usage practices, yet privacy-by-design principles still appear secondary in product development 3.

Even after establishing dedicated Privacy and Data Use teams and simplifying privacy controls following past incidents, Meta launched its AI app with a design that made private conversations public by default, which was viewed as a fundamental user experience failure 4.