Discovering Differences in Strategic Behavior Between Humans and LLMs

2026-02-10

AI Summary

This study investigates the origins of behavioral differences between large language models (LLMs) and humans in strategic interactions. Addressing the limitations of traditional approaches that rely on predefined models, the work introduces AlphaEvolve—an open-ended program discovery framework that learns interpretable behavioral programs directly from experimental data in iterative games such as Rock-Paper-Scissors, without requiring prior assumptions. The findings reveal that state-of-the-art LLMs exhibit greater strategic depth than human players, uncovering structural divergences in their underlying strategy-generation mechanisms. By enabling direct inference of executable strategies from behavioral data, this research establishes a novel paradigm for analyzing human–machine strategic behavior and advances the integration of interpretable AI with behavioral game theory.

Technology Category

• Humans and AI: Human-Aware Planning and Behavior Prediction
• Cognitive Modeling & Cognitive Systems: Simulating Human Behavior
• Game Theory and Economic Paradigms: Behavioral Game Theory

Application Category

• Economics, Online Markets and Human Computation: Uses of LLMs and GenAI for marketplace design, bidding, and strategic interactions
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation

Abstract

As Large Language Models (LLMs) are increasingly deployed in social and strategic scenarios, it becomes critical to understand where and why their behavior diverges from that of humans. While behavioral game theory (BGT) provides a framework for analyzing behavior, existing models do not fully capture the idiosyncratic behavior of humans or black-box, non-human agents like LLMs. We employ AlphaEvolve, a cutting-edge program discovery tool, to directly discover interpretable models of human and LLM behavior from data, thereby enabling open-ended discovery of structural factors driving human and LLM behavior. Our analysis on iterated rock-paper-scissors reveals that frontier LLMs can be capable of deeper strategic behavior than humans. These results provide a foundation for understanding structural differences driving differences in human and LLM behavior in strategic interactions.

Problem

• strategic behavior
• Large Language Models
• behavioral game theory
• human-AI differences
• iterated games

Innovation

• AlphaEvolve
• strategic behavior
• Large Language Models
• behavioral game theory
• program discovery