By buld_online 
Meta’s new AI-powered chatbot app is facing mounting criticism after users were found to be unknowingly sharing sensitive and often deeply personal conversations publicly. The standalone Meta AI app, launched in April, includes a social feed-style “Discover” tab that showcases interactions between users and the chatbot, ranging from travel tips and recipes to disturbing admissions and confidential queries.
While conversations are not shared by default, the platform features a prominent “share” button that allows users to post their exchanges to the public feed. However, critics argue the interface fails to make it sufficiently clear when content is being published publicly, resulting in a wave of posts revealing names, locations, phone numbers, legal concerns, and medical conditions.
Some of the posts visible on the feed include people asking for advice on tax evasion, court-related character statements containing full names, and even details of potential corporate fraud. Others have shared medical issues such as skin rashes, bowel-related problems, and post-surgery recovery queries, many of which are traceable to public Instagram profiles linked to their Meta accounts.
In a company blog post, Meta stated that “nothing is shared to your feed unless you choose to post it.” It also noted that the AI experience becomes more personalised when it draws on user data from Meta’s platforms, including information from Instagram profiles.
Still, the public rollout of this feature without clear guardrails has left many questioning Meta’s judgment. Since its launch, the Meta AI app has been downloaded approximately 6.5 million times, according to app intelligence firm Appfigures.
With growing concern over how easily identifiable information can be exposed, experts say Meta may need to revisit its approach to AI interaction and its assumptions about how people want their conversations shared.