Your conversations with Meta AI are not private, according to Meta, and everyone can see them


Meta’s new standalone AI chatbot app is facing increasing criticism over user privacy, as people are discovering that private chats and personal content are sometimes appearing in the app’s public Discover feed—often unintentionally.

Here’s what’s happening:

What users are seeing

  • Posts in the Discover feed are revealing highly personal or sensitive content, such as:

    • Beauty routines

    • Letters containing private details

    • Questions about government services (e.g., IRCTC rules for spouses)

    • Even odd or humorous prompts like “Why do some farts stink more than others?”

Why this is happening

  • The app includes a "Share" button in chatbot conversations.

  • When tapped, it opens a preview screen and asks for confirmation before posting to the public feed.

  • However, the language and design of the share prompt don’t make it clear:

    • What exactly is being shared

    • Where it’s being posted

    • Who can see it

Meta’s response

  • A spokesperson stated:

    “Users have to actively tap the share or publish buttons before it shows up on the app’s Discover feed.”

  • The company insists that chats are private by default and that only shared posts become public.

  • But even then, the message shown — “Prompts you post are public and visible to everyone” — lacks clarity about what that actually means in practice.

Real risks

  • Many users are accidentally making private chats public due to unclear UX.

  • Some are also sharing photos with Meta AI for transformations (e.g., comic or cartoon versions) — and these images are ending up visible to everyone.

  • People logging in via public Instagram profiles may unknowingly link their identity to these posts.

Current state

  • We’ve verified that:

    • Content is not automatically posted to Discover.

    • It is only posted if a user manually shares it.

    • However, the sharing interface is vague, which can easily confuse users.

Bottom line

Meta AI isn’t automatically leaking private conversations — but the design and wording of the sharing feature make it easy for users to misunderstand what will be made public. With highly personal or sensitive content increasingly showing up in the public feed, concerns about privacy, consent, and transparency are growing.


 

buttons=(Accept !) days=(20)

Our website uses cookies to enhance your experience. Learn More
Accept !