Prompt Injection
Date: 2026-01-28
Difficulty: Beginner
Delivered By: Harry Winchester
Overview
This brand new session covers the vunerabilities of generative AI and how prompts can be used to uncover sensitive information or perform prohibited actions using LLMs. Since Large Language Models have an inability to distinguish from developers and users, users can craft prompts in certain ways to extract all sorts of unintentional behaviour.
Prerequisites
There are no prerequisites required to take part in this session. All you need is a web browser and a curious mind.