Hackers have increasingly exploited ChatGPT to enhance their cyberattack capabilities, according to a recent report by OpenAI.
The report details how threat actors, particularly from Iran and China, have utilized the AI chatbot for tasks like debugging malware and crafting spear-phishing emails.
One notable group, CyberAv3ngers, linked to Iran’s Islamic Revolutionary Guard Corps, has targeted U.S. water utilities by leveraging ChatGPT to gather information on industrial control systems.
Another Iranian group, Storm-0817, used the tool to develop malware aimed at Android devices and create scripts for data theft.
The Chinese group SweetSpecter attempted to breach OpenAI employees’ accounts with malicious emails disguised as support requests.
Despite these activities, OpenAI asserts that hackers have not gained significant new capabilities through their use of AI, as many techniques remain achievable with existing tools.
OpenAI continues to monitor and disrupt these operations while enhancing its security measures against such abuses.
