PhD in Artificial Intelligence

This project will investigate attacks on large language models (LLMs)./nAssess the resistance of these models against new attacks using techniques from AI and optimisation, and develop methods to defend against such harms by leveraging cryptographic approaches./nKey tasks include adapting existing adversarial attacks for image classification and other domains against open source LLMs, investigating data poisoning attacks, performing attacks under different scenarios and modelling the impact, evaluating attacks in multi-agentic systems, and designing new defence methods that limit computational overhead to minimise energy and environmental cost.