Thesis Title
Machine learning and discrimination in foundation models

Research Description
Sadia’s research project aims to understand the impact of bias in the Generative AI foundation model on society and how Generative AI bias reproduces and amplifies structural inequality. Her main aim is to develop a framework to evaluate bias in Generative AI models and develop mitigation techniques by developing new metrics using computational tools. Her research will also focus on understanding whether it is possible to translate human rights primarily expressed to bind states into encouraging and binding private companies to address inequality. While doing this, she will address conflict over competing rights and the extent of commercial responsibility for systemic problems. Her research seeks to build improved methodologies to mitigate harmful bias and help companies build Responsible AI by providing legal, ethical, and technical solutions.

Supervisors
Professor Nic Suzor
Dr Kylie Pappalardo