People expect less but show equal influence of LLM versus human moral advice
Computers in Human Behavior Reports · 2026-12 · 24 · 101413
DOI: 10.1016/j.chbr.2026.101413 ↗Kurz gefasst
Fachartikel in der Zeitschrift Computers in Human Behavior Reports, 2026-12. Verfasst von Neele Engelmann und 4 weiteren Autor:innen. Die Arbeit nennt 48 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
People are often skeptical of AI in the moral domain, yet AI-generated ethical advice can influence moral judgments as much as human advice. Unlike most prior work, which attaches human or AI labels to identical advice, we examined how much real advice written by humans and by AI influences moral judgments, and whether people expect less influence from AI than from human advice. First, we developed 24 realistic everyday moral scenarios and collected over 500 persuasive advice texts from both incentivized human writers (N = 92) and GPT-4o. AI-generated advice was longer, more linguistically complex, more semantically uniform, more virtue-oriented, and less extreme than human advice. Next, a new sample of participants (N = 390) either revised initial moral judgments after receiving human or AI advice or indicated how much they expected such advice to influence them. Participants were equally influenced by human and AI moral advice, revising their judgments by a little under a third of the distance to the advice. Participants who only estimated the influence, however, expected less influence from AI than from human advice. These findings suggest that skepticism toward AI in moral contexts may shape people's expectations about AI moral advice, but not how much they are actually influenced by it.
Zitieren
APA
Engelmann, N., Schmidt, E., Alhosseini, A., Hannikainen, I. R., & Dong, M. (2026). People expect less but show equal influence of LLM versus human moral advice. Computers in Human Behavior Reports, 24, 101413. https://doi.org/10.1016/j.chbr.2026.101413
Vancouver
Engelmann N, Schmidt E, Alhosseini A, Hannikainen IR, Dong M. People expect less but show equal influence of LLM versus human moral advice. Computers in Human Behavior Reports. 2026;24:101413. doi:10.1016/j.chbr.2026.101413
BibTeX
@article{Engelmann2026,
author = {Engelmann, Neele and Schmidt, Eva-Madeleine and Alhosseini, Ali and Hannikainen, Ivar Rodríguez and Dong, Mengchen},
title = {People expect less but show equal influence of LLM versus human moral advice},
journal = {Computers in Human Behavior Reports},
year = {2026},
volume = {24},
pages = {101413},
doi = {10.1016/j.chbr.2026.101413}
}
EUVIMED