Navigating Advice Landscapes: A Comparative Analysis of Human-Human and Human-Chatbot Interactions Utilizing Large Language Models

Recipient: Sebastian Reinhardt, majoring in Software Engineering

Project Overview: The first aim of this research project is to compare and contrast the quality of advice that is given as a result of human-human interactions compared to the advice obtained through human-chat bot interactions, using Large Language Models (LLMs), such as ChatGPT (OpenAI). Through this, it will become clear whether which advice people prefer and trust more. The second aim of this project is to examine the implicit biases of the LLMs, in terms of the assumptions made, based off the information given in the prompt.

SUPERVISORS

Golnoosh Farnadi is an assistant professor at the School of Computer Science, McGill University, and an adjunct professor at Université de Montréal. She is a core academic member of Mila – Quebec Artificial Intelligence Institute and holds a Canada CIFAR AI Chair. Farnadi founded and is a principal investigator of the EQUAL lab at Mila / McGill University. The EQUAL lab (EQuity & EQuality Using AI and Learning algorithms) is a cutting-edge research laboratory dedicated to advancing the fields of algorithmic fairness and responsible AI.

Ross Otto is an Associate Professor at the Department of Psychology of McGill University. His research areas are in Behavioral Neuroscience and Cognition & Cognitive Neuroscience. His research investigates why we sometimes rely on slow and effortful choices, while at other times we rely on fast and reflexive choice, and answers such questions with the use of a combination of computational, behavioral, and psychophysiological, and neuroimaging techniques.