How NOT to Train Your Hack Bot - Dos and Don'ts of Building Offensive GPTs

Explore the potential and limitations of using large language models (LLMs) like GPT4 for offensive security operations in this 36-minute Black Hat conference talk. Discover techniques for leveraging LLMs to identify security vulnerabilities in applications, delve into the inner workings of these models, and learn state-of-the-art methods for their application in offensive contexts. Gain insights from speakers Shane Caldwell and Ariel Herbert-Voss as they demonstrate both successful and unsuccessful approaches to utilizing LLMs in cybersecurity, providing a comprehensive overview of the dos and don'ts when building offensive GPTs.