Integrace umělé inteligence v problematice penetračního testování: trendy, výhody a výzvy

Abstract

This thesis focuses on the integration of artificial intelligence into penetration testing of web applications, emphasizing current trends, benefits, and challenges of this approach. Its goal is to analyze the potential of large language models for automating vulnerability detection, compare available pre-trained models, and evaluate their effectiveness. The study introduces an experimental environment in which the capabilities of these tools to identify various types of security threats are tested, while also discussing their advantages and limitations in practical penetration testing. A key outcome of this work is the development of a custom tool that leverages artificial intelligence for automating vulnerability detection in web applications.