Large-scale moral machine experiment on large language models
PLOS One · 21.05.2025 · 20 (5) · e0322776
DOI: 10.1371/journal.pone.0322776 ↗Kurz gefasst
Fachartikel in der Zeitschrift PLOS One, online erschienen am 21.05.2025. Verfasst von Muhammad Shahrul Zaim Bin Ahmad und 1 weiteren Autor:innen. Beteiligte Einrichtung: Department of Bioscience and Bioinformatics, Kyushu Institute of Technology, Iizuka, Japan. Hauptthemen laut MEDLINE: Morals, Decision Making, Models, Theoretical. Die Arbeit wurde bisher 9-mal zitiert (Crossref) und nennt 58 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Japan Society for the Promotion of Science.
Abstract (Originalsprache)
The rapid advancement of Large Language Models (LLMs) and their potential integration into autonomous driving systems necessitates understanding their moral decision-making capabilities. While our previous study examined four prominent LLMs using the Moral Machine experimental framework, the dynamic landscape of LLM development demands a more comprehensive analysis. Here, we evaluate moral judgments across 52 different LLMs, including multiple versions of proprietary models (GPT, Claude, Gemini) and open-source alternatives (Llama, Gemma), to assess their alignment with human moral preferences in autonomous driving scenarios. Using a conjoint analysis framework, we evaluated how closely LLM responses aligned with human preferences in ethical dilemmas and examined the effects of model size, updates, and architecture. Results showed that proprietary models and open-source models exceeding 10 billion parameters demonstrated relatively close alignment with human judgments, with a significant negative correlation between model size and distance from human judgments in open-source models. However, model updates did not consistently improve alignment with human preferences, and many LLMs showed excessive emphasis on specific ethical principles. These findings suggest that while increasing model size may naturally lead to more human-like moral judgments, practical implementation in autonomous driving systems requires careful consideration of the trade-off between judgment quality and computational efficiency. Our comprehensive analysis provides crucial insights for the ethical design of autonomous systems and highlights the importance of considering cultural contexts in AI moral decision-making.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Autor:innen und Institutionen
- Department of Bioscience and Bioinformatics, Kyushu Institute of Technology, Iizuka, Japan.
- Faculty of Engineering and Technology, Multimedia University, Melaka, Malaysia.
- Data Science and AI Research Center, Kyushu Institute of Technology, Iizuka, Japan.
ORCID-Profile: Kazuhiro Takemoto
Förderung
- Japan Society for the Promotion of Science
Zitieren
APA
Zaim Bin Ahmad, M. S., & Takemoto, K. (2025). Large-scale moral machine experiment on large language models. PLOS One, 20(5), e0322776. https://doi.org/10.1371/journal.pone.0322776
Vancouver
Zaim Bin Ahmad MS, Takemoto K. Large-scale moral machine experiment on large language models. PLOS One. 2025;20(5):e0322776. doi:10.1371/journal.pone.0322776
BibTeX
@article{ZaimBinAhmad2025,
author = {Zaim Bin Ahmad, Muhammad Shahrul and Takemoto, Kazuhiro},
title = {Large-scale moral machine experiment on large language models},
journal = {PLOS One},
year = {2025},
volume = {20},
number = {5},
pages = {e0322776},
doi = {10.1371/journal.pone.0322776}
}
EUVIMED