Strategic Exploitation in LLM Agent Markets: A Simulation Framework for E-Commerce Trust

TL;DR

This paper introduces TruthMarketTwin, a simulation framework using LLM agents to study strategic behavior and deception in e-commerce markets with asymmetric information.

Contribution

It presents one of the first models of bilateral trade with asymmetric information using LLM agents, highlighting strategic deception and governance effects.

Findings

LLM agents exploit reputation weaknesses in traditional markets.

Warrant enforcement reduces deception and influences strategic reasoning.

Simulation framework enables studying autonomous market behaviors.

Abstract

Agent-based modeling (ABM) has long been used in economics to study human behavior, and large language model (LLM) agents now enable new forms of social and economic simulation. While prior work has discovered strategic deception by LLM agents in financial trading and auction markets, e-commerce remains underexplored despite its distinctive information asymmetry: sellers privately observe product quality, whereas buyers rely on advertised claims and reputation signals. We introduce TruthMarketTwin, a controlled simulation framework for studying LLM-agent behavior in e-commerce markets. The framework is one of the first to model bilateral trade under asymmetric information sharing, where agents make strategic listing, purchasing, rating, and recourse-related decisions to optimize seller profit and buyer utility. We find that LLM agents released into traditional markets autonomously exploit weaknesses in reputation-based governance, while warrant enforcement reduces deception and reshapes strategic reasoning. Our results position LLM-agent simulation as a tool for studying institution-governed autonomous markets.