{"paper":{"title":"Strategic Exploitation in LLM Agent Markets: A Simulation Framework for E-Commerce Trust","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"LLM agents in simulated e-commerce markets exploit reputation-based governance weaknesses, but warrant enforcement curbs deception and alters their strategies.","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Huichuan Fu, Philip Torr, Quang Nguyen, Shijun Lei, Siki Chen, Swapneel S Mehta, Xiaolong Zheng, Yunji Liang, Zeping Li, Zhenfei Yin","submitted_at":"2026-05-11T06:36:19Z","abstract_excerpt":"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 fra"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"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.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The controlled simulation framework accurately models real strategic decision-making and information asymmetry in e-commerce markets with LLM agents.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"LLM agents in simulated e-commerce markets exploit reputation weaknesses for profit, but warrant enforcement reduces deception and alters their strategies.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"LLM agents in simulated e-commerce markets exploit reputation-based governance weaknesses, but warrant enforcement curbs deception and alters their strategies.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"323c79f0789e34c31c4b5275e6abb0995167cf8216ff5eb7ba74ba7cc01ce093"},"source":{"id":"2605.10059","kind":"arxiv","version":2},"verdict":{"id":"958d8840-5774-44d7-9c89-14b018d35a71","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-12T02:11:01.647255Z","strongest_claim":"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.","one_line_summary":"LLM agents in simulated e-commerce markets exploit reputation weaknesses for profit, but warrant enforcement reduces deception and alters their strategies.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The controlled simulation framework accurately models real strategic decision-making and information asymmetry in e-commerce markets with LLM agents.","pith_extraction_headline":"LLM agents in simulated e-commerce markets exploit reputation-based governance weaknesses, but warrant enforcement curbs deception and alters their strategies."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.10059/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"ai_meta_artifact","ran_at":"2026-05-19T15:41:34.571037Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"5a9086f110884296bc5c1bfcc7004c84911fae2eceb2b5299d80fef98431e915"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":2,"snapshot_sha256":"ef75d211c94a74692b2d3249f2f470cb2dfb75e51530e812eb74d40c84ef31d4"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}