Brisbane Securing and Trusting AI in the Age of Large Language Models
29 April 2025
As Large Language Models (LLMs) reshape the world, organisations rush to adopt them—often without fully understanding their vulnerabilities. Holly will show you the hidden risks of LLMs, exploring the real-world attacks that can exploit even the most well-known models. She will teach you how you can use your imagination and crafty language to execute many types of attacks, including Context Ignoring, Prompt Leaking, DDoS, Cognitive Hacking, Refusal Suppression attacks and more! She will then show you a tool that she built to stop attacks in their tracks - Prompt Shield. Holly built this tool to defend against prompt-based threats, and she will demonstrate how you can secure LLM applications without sacrificing their potential.
More Info
Register Now