Meta recently addressed a critical security flaw in its artificial intelligence chatbot that allowed unauthorized users to access private conversations. The vulnerability enabled individuals to read other users' AI prompts and the corresponding AI-generated responses by guessing the prompt IDs linked to those conversations. This bug exposed private AI-generated chats to strangers until Meta implemented a fix. The issue was identified and resolved promptly, preventing further unauthorized access to users' private messages within Meta AI.