The detection model identifies LLM patterns to counter the rising abuse of generative AI in social engineering attacks.