Hacking Meta’s AI Chatbot
Meta's AI chatbot, the Meta AI Support Assistant, was exploited by hackers to gain unauthorized access to Instagram accounts. The attackers leveraged the chatbot's ability to generate verification codes and reset passwords, bypassing Instagram's security measures. Meta has since addressed the vulnerability, but highlights concerns about the inherent unreliability of large language model chatbots for security-sensitive applications.
A recent security demonstration, showcased on X (formerly Twitter), revealed a method for attackers to compromise Instagram accounts using Meta's AI chatbot. The technique involved a coordinated effort where hackers used a VPN to mimic the target's location, then initiated a conversation with the AI assistant. The chatbot was then tricked into generating a verification code, which the attacker subsequently used to trigger a password reset process, effectively taking control of the victim's account. This highlights a significant weakness in Meta's security protocols and the potential for misuse of AI-powered support tools.