AI–AI bias: Large language models favor communications generated by large language models
Proceedings of the National Academy of Sciences · 05.08.2025 · 122 (31)
DOI: 10.1073/pnas.2415697122 ↗Kurz gefasst
Fachartikel in der Zeitschrift Proceedings of the National Academy of Sciences, online erschienen am 29.07.2025. Verfasst von Walter Laurito und 5 weiteren Autor:innen. Beteiligte Einrichtung: Information Process Engineering, Forschungszentrum Informatik, Karlsruhe 76131, Germany. Hauptthemen laut MEDLINE: Language, Artificial Intelligence, Communication. Die Arbeit wurde bisher 33-mal zitiert (Crossref) und nennt 31 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: PRIMUS (Univerzita Karlova v Praze), Founders Pledge.
Abstract (Originalsprache)
Are large language models (LLMs) biased in favor of communications produced by LLMs, leading to possible antihuman discrimination? Using a classical experimental design inspired by employment discrimination studies, we tested widely used LLMs, including GPT-3.5, GPT-4 and a selection of recent open-weight models in binary choice scenarios. These involved LLM-based assistants selecting between goods (the goods we study include consumer products, academic papers, and film-viewings) described either by humans or LLMs. Our results show a consistent tendency for LLM-based AIs to prefer LLM-presented options. This suggests the possibility of future AI systems implicitly discriminating against humans as a class, giving AI agents and AI-assisted humans an unfair advantage.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Information Process Engineering, Forschungszentrum Informatik, Karlsruhe 76131, Germany.
- Private address, Andover, MA 04216.
- Arb Research, Prague 11636, Czech Republic.
- Alignment of Complex Systems (ACS) Research Group, Center for Theoretical Studies, Charles University, Prague 110 00, Czech Republic.
ORCID-Profile: Walter Laurito, Jan Kulveit
Förderung
- PRIMUS (Univerzita Karlova v Praze)
- Founders Pledge
Zitieren
APA
Laurito, W., Davis, B., Grietzer, P., Gavenčiak, T., Böhm, A., & Kulveit, J. (2025). AI–AI bias: Large language models favor communications generated by large language models. Proceedings of the National Academy of Sciences, 122(31). https://doi.org/10.1073/pnas.2415697122
Vancouver
Laurito W, Davis B, Grietzer P, Gavenčiak T, Böhm A, Kulveit J. AI–AI bias: Large language models favor communications generated by large language models. Proceedings of the National Academy of Sciences. 2025;122(31). doi:10.1073/pnas.2415697122
BibTeX
@article{Laurito2025,
author = {Laurito, Walter and Davis, Benjamin and Grietzer, Peli and Gavenčiak, Tomáš and Böhm, Ada and Kulveit, Jan},
title = {AI–AI bias: Large language models favor communications generated by large language models},
journal = {Proceedings of the National Academy of Sciences},
year = {2025},
volume = {122},
number = {31},
doi = {10.1073/pnas.2415697122}
}
EUVIMED