US12423441B2 - Method for Using Generative Large Language Models (LLM) for Cybersecurity Deception and Honeypots