Can Machines Think Like Humans? A Behavioral Evaluation of LLM Agents in Dictator Games

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autor principal: Ma, Ji
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866909909063303168
author Ma, Ji
author_facet Ma, Ji
contents As Large Language Model (LLM)-based agents increasingly engage with human society, how well do we understand their prosocial behaviors? We (1) investigate how LLM agents' prosocial behaviors can be induced by different personas and benchmarked against human behaviors; and (2) introduce a social science approach to evaluate LLM agents' decision-making. We explored how different personas and experimental framings affect these AI agents' altruistic behavior in dictator games and compared their behaviors within the same LLM family, across various families, and with human behaviors. The findings reveal that merely assigning a human-like identity to LLMs does not produce human-like behaviors. These findings suggest that LLM agents' reasoning does not consistently exhibit textual markers of human decision-making in dictator games and that their alignment with human behavior varies substantially across model architectures and prompt formulations; even worse, such dependence does not follow a clear pattern. As society increasingly integrates machine intelligence, "Prosocial AI" emerges as a promising and urgent research direction in philanthropic studies.
format Preprint
id arxiv_https___arxiv_org_abs_2410_21359
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Can Machines Think Like Humans? A Behavioral Evaluation of LLM Agents in Dictator Games
Ma, Ji
Computation and Language
Artificial Intelligence
Computers and Society
Machine Learning
General Economics
Economics
As Large Language Model (LLM)-based agents increasingly engage with human society, how well do we understand their prosocial behaviors? We (1) investigate how LLM agents' prosocial behaviors can be induced by different personas and benchmarked against human behaviors; and (2) introduce a social science approach to evaluate LLM agents' decision-making. We explored how different personas and experimental framings affect these AI agents' altruistic behavior in dictator games and compared their behaviors within the same LLM family, across various families, and with human behaviors. The findings reveal that merely assigning a human-like identity to LLMs does not produce human-like behaviors. These findings suggest that LLM agents' reasoning does not consistently exhibit textual markers of human decision-making in dictator games and that their alignment with human behavior varies substantially across model architectures and prompt formulations; even worse, such dependence does not follow a clear pattern. As society increasingly integrates machine intelligence, "Prosocial AI" emerges as a promising and urgent research direction in philanthropic studies.
title Can Machines Think Like Humans? A Behavioral Evaluation of LLM Agents in Dictator Games
topic Computation and Language
Artificial Intelligence
Computers and Society
Machine Learning
General Economics
Economics
url https://arxiv.org/abs/2410.21359