-
Using Large Language Models to Simulate and Predict Human Decision-Making
We explore how large language models can be used to predict human decisions in language-based persuasion games, comparing direct prompting, LLM-based data generation, and hybrid methods that mix synthetic and human data.
-
Towards LLM-based Experimental Economics
This blog post explores how Large Language Models (LLMs) can revolutionize experimental economics by simulating human behavior in foundational economic scenarios, while steering AI from zero-sum games to general non-cooperative games and agents towards strategic entities in a multi-agent environment. Traditional approaches in economics are limited by the costs and complexities of human-based experiments, and replacing humans with LLMs enables large-scale simulations, allowing researchers to examine a variety of economic mechanisms.