{"version":1,"type":"story","url":"https://digestai.news/story/open-source-llms-over-cooperate-in-prisoner-s-dilemma-study-shows","json":"https://digestai.news/story/open-source-llms-over-cooperate-in-prisoner-s-dilemma-study-shows.json","markdown":"https://digestai.news/story/open-source-llms-over-cooperate-in-prisoner-s-dilemma-study-shows.md","slug":"open-source-llms-over-cooperate-in-prisoner-s-dilemma-study-shows","headline":"Open-source LLMS over-cooperate in prisoner's dilemma, study shows","summary":"The paper reports that open‑source large language models (LLMs) tend to cooperate far more than humans in repeated Prisoner’s Dilemma games. Researchers tested six models—TII Falcon 7B Instruct, Google Gemma 7B, Meta Llama 3.1 8B Instruct, Microsoft Phi Mini MoE Instruct, and Qwen 2.5 7B Instruct—using a standardized prompting framework.\n\nIn a shuffled‑partner setting, human cooperation averaged 28.5 % over 100 rounds, while several LLMs exceeded 80 %. Qwen‑2.5‑7B‑Instruct behaved deterministically, whereas Phi‑Mini‑MoE‑Instruct showed greater variability and briefly matched human rates. The study also examined a fixed‑partner setting where both agents retained memory.\n\nThese results suggest that current open‑source LLMs are not yet reliable substitutes for human participants in behavioral simulations, highlighting the need for better memory mechanisms and more realistic decision‑making models.","keyPoints":["Open‑source llms achieved cooperation rates above 80% in shuffled‑partner setting","Human cooperation averaged 28.5% in same setting","Qwen‑2.5‑7B‑Instruct behaved deterministically, Phi‑Mini‑MoE‑Instruct variable"],"whyItMatters":"The findings expose limits of current LLMs for modeling human decision making, underscoring the need for improved memory and realism in AI agents used for social simulations.","category":{"slug":"research","name":"Research","url":"https://digestai.news/category/research"},"entities":{"companies":["TII","Google","Meta","Microsoft","Qwen"],"models":["TII Falcon 7B Instruct","Google Gemma 7B","Meta Llama 3.1 8B Instruct","Microsoft Phi Mini MoE Instruct","Qwen 2.5 7B Instruct"],"people":["Montero-Porras"]},"firstPublishedAt":"2026-10-20T04:00:00Z","updatedAt":"2026-10-20T04:00:00Z","sourceCount":1,"hasPrimarySource":true,"sources":[{"outlet":"arXiv cs.CL","title":"Modeling Human Behavior in the Prisoner's Dilemma with Large Language Models","url":"https://arxiv.org/abs/2610.23860","publishedAt":"2026-10-20T04:00:00Z","type":"primary","primary":true,"lead":true}],"sourceNotes":null,"discussions":[],"thread":null,"cite":{"text":"Digest AI, \"Open-source LLMS over-cooperate in prisoner's dilemma, study shows\", 20 October 2026, https://digestai.news/story/open-source-llms-over-cooperate-in-prisoner-s-dilemma-study-shows","publisher":"Digest AI","title":"Open-source LLMS over-cooperate in prisoner's dilemma, study shows","datePublished":"2026-10-20T04:00:00Z","url":"https://digestai.news/story/open-source-llms-over-cooperate-in-prisoner-s-dilemma-study-shows"},"generatedBy":"Written by Digest AI's editorial model from the linked sources; the sources are the record.","license":"Headlines, digests and key points are written by Digest AI and may be quoted with a link to the story page. Linked articles belong to their publishers. Terms: https://digestai.news/terms#reuse"}