A new version of this entry is available:
Loading...
Article
2026
Does a smarter ChatGPT become more utilitarian?
Does a smarter ChatGPT become more utilitarian?
Abstract (English)
Hundreds of millions of users interact with large language models (LLMs) regularly to get advice on all aspects of life. The increase in LLMs’ logical capabilities might be accompanied by unintended side effects with ethical implications. Focusing on recent model developments of ChatGPT, we can show clear evidence for a systematic shift in ethical stances that accompanied a leap in the models’ logical capabilities. Specifically, as ChatGPT’s capacity grows, it tends to give decisively more utilitarian answers to the two most famous dilemmas in ethics. Given the documented impact that LLMs have on users, we call for a research focus on the prevalence and dominance of ethical theories in LLMs as well as their potential shift over time. Moreover, our findings highlight the need for continuous monitoring and transparent public reporting of LLMs’ moral reasoning to ensure their informed and responsible use.
File is subject to an embargo until
This is a correction to:
A correction to this entry is available:
This is a new version of:
Other version
Notes
Publication license
Publication series
Published in
Science and engineering ethics, 32 (2026), 1, 1.
https://doi.org/10.1007/s11948-025-00579-4.
ISSN: 1471-5546
Dordrecht : Springer Netherlands
Other version
Faculty
Institute
Examination date
Supervisor
Cite this publication
Pfeffer, J., Krügel, S., & Uhl, M. (2026). Does a Smarter ChatGPT Become More Utilitarian? Science and engineering ethics, 32(1). https://doi.org/10.1007/s11948-025-00579-4
Edition / version
Citation
DOI
ISSN
ISBN
Language
English
Publisher
Publisher place
Classification (DDC)
170 Ethics
Original object
University bibliography
Standardized keywords (GND)
Sustainable Development Goals
BibTeX
@article{Pfeffer2026,
doi = {10.1007/s11948-025-00579-4},
author = {Pfeffer, Jürgen and Krügel, Sebastian and Uhl, Matthias et al.},
title = {Does a Smarter ChatGPT Become More Utilitarian?},
journal = {Science and Engineering Ethics},
year = {2026},
volume = {32},
number = {1},
}