Events

This event will also take place online.

• More information about the Talking to Machines Seminar Series and how to attend online

Abstract

Large Language Models (LLMs) are a transformational technology, fundamentally changing how people obtain information and interact with the world. As people become increasingly reliant on them for an enormous variety of tasks, a body of academic research has developed to examine these models for inherent biases, especially political biases, often finding them small.

We challenge this prevailing wisdom. First, by comparing 31 LLMs to legislators, judges, and a nationally representative sample of U.S. voters, we show that LLMs’ apparently small overall partisan preference is the net result of offsetting extreme views on specific topics, much like moderate voters. Second, in a randomized experiment, we show that LLMs can promulgate their preferences into political persuasiveness even in information-seeking contexts: voters randomized to discuss political issues with an LLM chatbot are as much as 5 percentage points more likely to express the same preferences as that chatbot. 

Speaker

Aaron R Kaufman
Associate Professor of Political Science

Aaron Kaufman is Associate Professor of Political Science at NYU Abu Dhabi, and Co-Director of the Center for Interdisciplinary Data Science and Artificial Intelligence. His work applies computational tools to measurement problems in political science, including ideology, discrimination, policy significance, and legislative district compactness. His work has appeared in Nature, Nature Scientific Data, Nature Scientific Reports, the APSR, AJPS, BJPS, JOP, Political Analysis, the British Medical Journal, and the Journal of Quantitative Analysis in Sports. He received his PhD in Political Science and AM in Statistics from Harvard, and his BA in Political Science from the University of California, Berkeley.