Biosketch
Molly obtained her Bachelor’s degree in Modern and Medieval Languages at the University of Cambridge. She then completed a MSc in Data Science and Artificial Intelligence at Queen Mary University of London. On completion of her MSc she worked as a data engineer at First Derivative. Since May 2023 she is a relAI member.
relAI Research
Beyond the Black Box: Explainable AI for Understanding and Mitigating Media Bias in Large Language Models
The increasing reliance on large language models (LLMs) for news generation and media analysis raises concerns about bias in AI-generated content. While LLMs offer powerful tools for automated text generation, their outputs often reflect biases inherent in their training data, fine-tuning processes, or alignment mechanisms. Explainable AI (XAI) provides a promising avenue for making AI-generated content more transparent, interpretable, and accountable by shedding light on how these models develop and express biases.
This research aims to explore XAI methodologies to detect, interpret, and mitigate media bias in LLMs.