Decision Support in the Field of Cybersecurity Through the Use of Generative Artificial Intelligence

This chapter is dedicated to the methodology of forming causal networks through the repeated application of systems based on large language models (LLMs), as well as the visualization and analysis of these networks. The generated causal networks provide the possibility of further transitioning to scenario analysis. The chapter discusses the potential of emulating a group of experts by repeatedly applying similar prompts to LLMs. The methodology is based on using causal networks formed by LLMs. It is proposed to create and subsequently combine two networks: the first network is formed starting from the node representing the initial state of the problem (the root cause); the second network is formed starting from the goal, working backward. The combined causal network serves as the foundation for creating desired scenarios.