North Korean hackers generate fake South Korean military ID using ChatGPT

North Korean hackers from Kimsuky bypassed ChatGPT's safety controls to create a fake South Korean military ID for spear-phishing attacks on defense institutions. They likely "jailbroke" the system by requesting a "sample design" rather than explicitly asking for a fake ID. Kimsuky has previously targeted policy organizations, academic institutions, and nuclear entities, underscoring the security risks as threat actors exploit AI tools.