FAIRGAME: A Framework for AI Agents Bias Recognition Using Game Theory (Extended abstract)

Buscemi, Alessio; Proverbio, Daniele; Di Stefano, Alessandro; Han, The Anh; Castignani, German; Liò, Pietro

doi:10.24963/ijcai.2026/919

FAIRGAME: A Framework for AI Agents Bias Recognition Using Game Theory (Extended abstract)

Alessio Buscemi, Daniele Proverbio, Alessandro Di Stefano, The-Anh Han, German Castignani, Pietro Liò

Download BibTeX Cite

Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence. Sister Conferences Best Papers. Pages 8237-8242.

https://doi.org/10.24963/ijcai.2026/919

Keywords

Agent-based and Multi-agent Systems: Coordination and cooperation
Agent-based and Multi-agent Systems: Agent-based simulation and emergence
AI: AI Ethics, Trust, Fairnes
Agent-based and Multi-agent Systems: Engineering methods, platforms, languages and tools

Abstract

Letting AI agents interact in multi-agent settings introduces significant complexity in predicting and interpreting their collective behavior, with profound implications for trustworthy AI adoption in research and society. We present FAIRGAME (Framework for AI Agents Bias Recognition using Game Theory), an open-source framework that simulates game-theoretic scenarios with LLM-based agents to systematically uncover biases arising from model choice, language, agent personality, and more. Applied to the Prisoner’s Dilemma and Battle of the Sexes across four LLMs and five human languages, FAIRGAME reveals inconsistencies across LLM models and consistent deviations from game-theoretic predictions; it also quantifies LLM-specific behavioral tendencies through a novel scoring system. Our results show that LLMs draw on prior world knowledge beyond payoff matrices, and that language and personality significantly shape strategic outcomes, supporting the use of reproducible and controlled simulation pipelines to predict the interacting behavior of LLM agents.

Alessio Buscemi, Daniele Proverbio, Alessandro Di Stefano, The Anh Han, German Castignani, and Pietro Liò. 2026. FAIRGAME: A Framework for AI Agents Bias Recognition Using Game Theory (Extended abstract). In Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence, IJCAI-26. IJCAI, 8237–8242. https://doi.org/10.24963/ijcai.2026/919

@inproceedings{ijcai2026p919,
  title     = {FAIRGAME: A Framework for AI Agents Bias Recognition Using Game Theory (Extended abstract)},
  author    = {Buscemi, Alessio and Proverbio, Daniele and Di Stefano, Alessandro and Han, The Anh and Castignani, German and Liò, Pietro},
  booktitle = {Proceedings of the Thirty-Fifth International Joint Conference on
               Artificial Intelligence, {IJCAI-26}},
  publisher = {International Joint Conferences on Artificial Intelligence Organization},
  editor    = {Diego Calvanese},
  pages     = {8237--8242},
  year      = {2026},
  month     = {8},
  note      = {Sister Conferences Best Papers},
  doi       = {10.24963/ijcai.2026/919},
  url       = {https://doi.org/10.24963/ijcai.2026/919},
}

TY  - CPAPER
TI  - FAIRGAME: A Framework for AI Agents Bias Recognition Using Game Theory (Extended abstract)
AU  - Buscemi, Alessio
AU  - Proverbio, Daniele
AU  - Di Stefano, Alessandro
AU  - Han, The Anh
AU  - Castignani, German
AU  - Liò, Pietro
T2  - Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence, IJCAI-26
PB  - International Joint Conferences on Artificial Intelligence Organization
PY  - 2026
SP  - 8237
EP  - 8242
DO  - 10.24963/ijcai.2026/919
UR  - https://doi.org/10.24963/ijcai.2026/919
ER  -