Populism Meets AI: Advancing Populism Research with LLMs

• In Event: New Methods in Political Communication

Brief Overview

The study demonstrates that LLMs can match expert human coders in detecting populist discourse, achieving near-perfect agreement scores, proving that AI can now grade the moral rhetoric of “people versus elites” as reliably as trained human coders.

Authors

• Yujin Julia Jung, Mount St. Mary's University

• Eduardo Ryo Tamaki, German Institute of Global Area Studies

• Kirk A. Hawkins, Brigham Young University, Provo

• Levente Littvay, Centre for Social Sciences, Budapest