On the usage of ChatGPT for integrating CAPEC attacks into ADVISE Meta ontology

In today’s cybersecurity landscape, robust security assessment methodologies are essential for evaluating and improving systems, networks, applications, and data security. Modeling and simulation play an important role in this process by providing meaningful representations and analyses of attacks and defense strategies, particularly in systems where security breaches could have devastating consequences. The ADversary VIew Security Evaluation (ADVISE) Meta framework offers an ontology-based approach that, starting from a system’s architectural model, automatically generates detailed security models representing the attack steps that adversaries might take to achieve their goals. Manually extending the ADVISE Meta ontology with specific attack patterns is a challenging task that involves a deep understanding of the ontology, and its semantics. It also requires analyzing the attack paths to identify the necessary information in the ontology. To address this challenge we propose a methodology to facilitate the integration of attack patterns into the ADVISE Meta framework using ChatGPT. We focus on the Common Attack Pattern Enumeration and Classification (CAPEC) catalog by MITRE, a popular catalog with more than 500 attack patterns describing the common attributes and approaches used by adversaries to exploit known weaknesses in IT systems. ChatGPT is used as a support tool to interpret the descriptions of the attacks in the CAPEC catalog and systematically integrate the interpreted data into the ADVISE Meta ontology to generate the attack steps.