(PI06) (INCYBER) LLM evaluation through automated red teaming with Giskard's open source library

In this session, you'll discover how to improve the security of Large Language Models (LLMs). We'll show you how to use red-teaming techniques from cybersecurity to identify and evaluate vulnerabilities in LLM applications, ensuring their safety and reliability. Additionally, you'll learn how Giskard's tools can be integrated into your workflow for automatic vulnerability detection, allowing you to scale your security efforts for Generative AI.