Large Language Models Assume People are More Rational than We Really are

TL;DR

Large Language Models tend to assume humans are more rational than they are, aligning more with expected value theory than actual human decision-making, which impacts their communication and prediction of human behavior.

Contribution

This paper reveals that state-of-the-art LLMs overestimate human rationality, aligning their models with expected value theory rather than actual human decision patterns.

Findings

LLMs assume higher rationality than humans exhibit

Models align more with expected value theory than real human behavior

Inferences about others' decisions are highly correlated between LLMs and humans

Abstract

In order for AI systems to communicate effectively with people, they must understand how we make decisions. However, people's decisions are not always rational, so the implicit internal models of human decision-making in Large Language Models (LLMs) must account for this. Previous empirical evidence seems to suggest that these implicit models are accurate -- LLMs offer believable proxies of human behavior, acting how we expect humans would in everyday interactions. However, by comparing LLM behavior and predictions to a large dataset of human decisions, we find that this is actually not the case: when both simulating and predicting people's choices, a suite of cutting-edge LLMs (GPT-4o & 4-Turbo, Llama-3-8B & 70B, Claude 3 Opus) assume that people are more rational than we really are. Specifically, these models deviate from human behavior and align more closely with a classic model of rational choice -- expected value theory. Interestingly, people also tend to assume that other people are rational when interpreting their behavior. As a consequence, when we compare the inferences that LLMs and people draw from the decisions of others using another psychological dataset, we find that these inferences are highly correlated. Thus, the implicit decision-making models of LLMs appear to be aligned with the human expectation that other people will act rationally, rather than with how people actually act.