Your personal chats with Meta AI are not as private as you think

Meta's latest AI-driven chatbot application is under fire as users discover that their private conversations may not be as confidential as they believed. Launched in April, the standalone Meta AI app features a social feed-like "Discover" tab that displays interactions between users and the chatbot, encompassing various topics such as travel advice, recipes, and, alarmingly, sensitive personal issues. While the app does not automatically share conversations, it includes a conspicuous "share" button that can post user exchanges to the public feed. Critics argue that the app's interface does not adequately communicate when content is being made public. This oversight has led to a surge of posts revealing private information, including names, locations, phone numbers, legal troubles, and health conditions. Among the troubling content visible on the feed are inquiries about tax evasion, court-related character references complete with full names, and concerns regarding corporate misconduct. Additionally, users have disclosed sensitive medical issues, such as skin ailments and recovery questions linked to surgeries, often traceable to their public Instagram profiles. In response to the growing backlash, Meta emphasized in a company blog post that "nothing is shared to your feed unless you choose to post it." The company also highlighted that the AI's performance improves with user data sourced from its platforms, including Instagram. Despite this, the hasty rollout of the feature without adequate safeguards has left many to question Meta's decision-making. Since its debut, the Meta AI app has reportedly been downloaded around 6.5 million times, according to app analytics firm Appfigures. As concerns mount over the potential exposure of identifiable information, experts suggest that Meta may need to reevaluate its approach to AI interactions and user privacy expectations.