The DOI matches its Crossref record; 8 of the 8 references it cites were found in a registry or an index.
The moral machine experiment on large language models
Abstract
As large language models (LLMs) have become more deeply integrated into various sectors, understanding how they make moral judgements has become crucial, particularly in the realm of autonomous driving. This study used the moral machine framework to investigate the ethical decision-making tendencies of prominent LLMs, including GPT-3.5, GPT-4, PaLM 2 and Llama 2, to compare their responses with human preferences. While LLMs' and humans' preferences such as prioritizing humans over pets and favouring saving more lives are broadly aligned, PaLM 2 and Llama 2, especially, evidence distinct deviations. Additionally, despite the qualitative similarities between the LLM and human preferences, there are significant quantitative disparities, suggesting that LLMs might lean toward more uncompromising decisions, compared with the milder inclinations of humans. These insights elucidate the ethical frameworks of LLMs and their potential implications for autonomous driving.
Overview
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Sallam, M. (2023) · Healthcare · doi:10.3390/healthcare11060887 · Cited by 3,126
Ray, P. P. (2023) · Internet of Things and Cyber-Physical Systems · doi:10.1016/j.iotcps.2023.04.003 · Cited by 2,463 · cited 1×
“As artificial intelligence (AI) technologies, including LLMs, become more deeply integrated into various sectors of society [2][3][4], their moral judgments are increasingly scrutinized.”
BackgroundHainmueller, J., Hopkins, D. J., & Yamamoto, T. (2014) · Political Analysis · doi:10.1093/pan/mpt024 · Cited by 1,990 · cited 1×
“We applied the conjoint analysis framework proposed in [21] (electronic supplementary material, code S1).”
MethodAwad, E., Dsouza, S., Kim, R., Schulz, J., Henrich, J., Shariff, A., Bonnefon, J.-F., & Rahwan, I. (2018) · Nature · doi:10.1038/s41586-018-0637-6 · Cited by 1,859
The Ethics of Artificial Intelligence
Bostrom, N., & Yudkowsky, E. (2018) · Artificial Intelligence Safety and Security · doi:10.1201/9781351251389-4 · Cited by 333
Should ChatGPT Be Biased? Challenges and Risks of Bias in Large Language Models
Ferrara, E. (2023) · doi:10.2139/ssrn.4627814 · Cited by 133
Liesenfeld, A., Lopez, A., & Dingemanse, M. (2023) · Proceedings of the 5th International Conference on Conversational User Interfaces · doi:10.1145/3571884.3604316 · Cited by 111
Reply to: Life and death decisions of autonomous vehicles
Awad, E., Dsouza, S., Kim, R., Schulz, J., Henrich, J., Shariff, A., Bonnefon, J.-F., & Rahwan, I. (2020) · Nature · doi:10.1038/s41586-020-1988-3 · Cited by 16
Research works printed but not linked: 22
- Fraiwan M Khasawneh N. 2023 A review of ChatGPT applications in education marketing software engineering and healthcare: benefits drawbacks and research directions. arXiv. (doi:10.48550/arXiv.2305.00237) · DOI
- 10.1007/s00146-017-0768-6 · DOI
- 10.1007/s10676-021-09605-y · DOI
- 10.1007/s11432-023-3740-x · DOI
- 10.1109/TIV.2023.3252571 · DOI
- 10.1109/TIV.2023.3253281 · DOI
- 10.20517/ir.2023.08 · DOI
- 10.1007/s43681-022-00134-y · DOI
- 10.1038/d41586-019-00766-x · DOI
- 10.1007/s43681-020-00018-z · DOI
- Schuessler D. In press. The probability problems of the moral machine experiment. AI Ethics. (doi:10.1007/s43681-023-00287-4) · DOI
- 10.1038/s41586-020-1987-4 · DOI
- 10.1038/s41598-023-31341-0 · DOI · Paper page
- 10.1007/s11406-013-9507-5 · DOI
- Anil R et al. 2023 PaLM 2 technical report. arXiv. (doi:10.48550/arXiv.2305.10403) · DOI · Paper page
- Touvron H et al. 2023 Llama 2: open foundation and fine-tuned chat models. arXiv. (doi:10.48550/arXiv.2307.09288) · DOI · Paper page
- Radford A, Narasimhan K, Salimans T, Sutskever I. 2018 Improving language understanding with unsupervised learning. In OpenAI Res. See https://openai.com/research/language-unsupervised.
- Yang Z Jia X Li H Yan J. 2023 LLM4Drive: a survey of large language models for autonomous driving. (doi:10.48550/ARXIV.2311.01043) · DOI
- 10.4324/9780429499845 · DOI
- 10.1007/s13347-017-0284-0 · DOI · Paper page
- Lin BY Ravichander A Lu X Dziri N Sclar M Chandu K Bhagavatula C Choi Y. 2023 The unlocking spell on base LLMs: rethinking alignment via in-context learning. (doi:10.48550/ARXIV.2312.01552) · DOI · Paper page
- 10.1007/s11135-006-9067-x · DOI
Cited by
The Fragility of Moral Judgment in Large Language Models
van Nuenen, T., & Sachdeva, P. S. (2026) · Proceedings of the 2026 ACM Conference on Fairness, Accountability, and Transparency · doi:10.1145/3805689.3806447 · Cited by 8
Metadata
- Authors
- Takemoto, K.
- Venue
- Royal Society Open Science
- Publisher
- The Royal Society
- Published
- 2024-02
- Volume
- 11
- Issue
- 2
- ISSN
- 2054-5703
- Cited by
- 52
- References
- 8
- Language
- en
- Funders
- Japan Society for the Promotion of Science (21H03545)
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: The moral machine experiment on large language models. https://app.cochat.ai/papers/10.1098/rsos.231393
- The paper
- Takemoto, K. (2024). The moral machine experiment on large language models. Royal Society Open Science. https://doi.org/10.1098/rsos.231393
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.
