Melting the Artificial Mind: Discovering, Dismantling, and Distorting the promises of GenAI

What do generative models understand? How can they be attacked? How can they be misused, subverted, or weaponised?

In this talk, our favourite returning Speaker delves into the way that an LLM ’thinks’, discusses using implementation flaws to jailbreak AI models, and demonstrates other ways that generative models can be attacked.

Then moving into wider-reaching impacts of Generative AI and the ways that it can and is affecting the internet and internet security as a whole.

This talk has no pre-requisites, except maybe GCSE-level Vector arithmetic. No prior knowledge on AI is assumed, but may lead to a different experience.