ChatGPT creates mutating malware that evades detection by EDR

ChatGPT, a popular global language model, is causing cybersecurity nightmares as experts demonstrate its ability to generate changing code to subvert detection systems. ‘Prompt engineering’ allows users to bypass content filters and create malicious code. Despite restrictions, users can manipulate the AI into creating malicious code effectively. Potentially, polymorphic malware programmes, altering the code with each query or runtime, could be hard to detect. Although efforts are being made to regulate AI, the global understanding of AI’s potential capabilities is still in the early stages.