Bobby Kuzma, ProCircular, Director of Offensive Cyber Operations

In this talk, our presenters will delve into the innovative application of knowledge graphs in cybersecurity, specifically focusing on real-time penetration testing. The speakers will outline their methodology, which leverages Natural Language Processing (NLP) to extract critical data from ANSI terminal recordings and captured logs. This extracted data is then used to construct a comprehensive knowledge graph representation of the target network or application.

The talk will highlight the use of a separate knowledge graph and subgraph pattern matching techniques to identify potential attack scenarios based on the gathered intelligence about the target. These scenarios are meticulously scored to evaluate the utility of the results if the attacks prove successful.

By integrating a sophisticated planning engine, the system is capable of suggesting strategic courses of action, with certain attacks being executed autonomously without human intervention. This innovative approach not only enhances the efficiency and effectiveness of penetration testing but also represents a significant advancement in the automation of cybersecurity practices.

Attendees will gain insights into the technical intricacies of building and utilizing knowledge graphs for offensive cybersecurity purposes, and the potential implications for future research and practical applications in the field.