Enabling Cybersecurity Education Through Digital Twins and Generative AI

In recent years, Artificial Intelligence (AI) is taking over many fields, changing the way activities and tasks are being carried out. Among the most recent AI-based technologies, Large Language Models (LLMs) and Digital Twins (DTs) are rapidly spreading, bringing advantages and innovation from multiple point of views. This research focuses on the employment of LLMs and DTs in cybersecurity, exploiting their ability to replicate infrastructures, allowing for real-time monitoring, threat analysis, and system simulation. Specifically, we investigate how integrating DTs and LLMs in penetration testing tools can enhance cybersecurity education and the effectiveness of interventions to face threats. With this perspective, we introduce a framework, called Cyber Red Team Knife (CRTK) for enhancing cybersecurity education with respect to vulnerabilities for both defensive and offensive strategies. The framework is based on DTs and Generative AI in order to support the penetration testing activities thanks to the Cyber Kill Chain model in order to automate penetration testing activities thanks to Red Team Knife (RTK) that integrates it. RTK is a custom penetration testing toolkit to guide learners through key phases of cyber-attacks within a DT-powered ecosystem.