Meta has suspended the conversation-sharing feature in its Muse AI assistant after reports that sharing could expose users' personal and business secrets. The move follows a privacy issue that surfaced when users shared their chats, according to Ars Technica.
The flaw meant that shared conversations could reveal sensitive information beyond what the user intended, including both personal and business data. This has raised concerns about how the assistant handles confidential content and whether its sharing controls are adequately protecting users.
Meta is investigating the issue and has paused the feature in the meantime. The incident highlights the broader risk of AI assistants that process and transmit sensitive user data, and the difficulty of ensuring that sharing mechanisms work as expected.