Meta Confirms Instagram Accounts Hacked via AI Chatbot Exploit

Meta revealed that a vulnerability in its AI chatbot enabled attackers to gain unauthorized access to thousands of Instagram accounts. The exploit involved manipulating the chatbot's natural language processing to execute commands that bypassed authentication controls, such as password resets or session token theft. Once compromised, accounts were used for spam and phishing campaigns. Meta has since patched the vulnerability and reset affected accounts, urging users to enable two-factor authentication. This incident highlights the emerging security risks of AI agents with elevated privileges, emphasizing the need for rigorous input validation and access controls in conversational AI systems.