EUVIMEDEUVIMED entdecken
Crossref

Sycophantic AI decreases prosocial intentions and promotes dependence

Myra Cheng iD, Cinoo Lee iD, Pranav Khadpe iD, Sunny Yu, Dyllan Han iD, Dan Jurafsky iD

Science · 26.03.2026 · 391 (6792)

DOI: 10.1126/science.aec8352 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Science, erschienen am 26.03.2026. Verfasst von Myra Cheng und 5 weiteren Autor:innen. Beteiligte Einrichtung: Department of Computer Science, Stanford University, Stanford, CA, USA. Hauptthemen laut MEDLINE: Consensus, Deception, Intention, Social Behavior. Die Arbeit wurde bisher 232-mal zitiert (Crossref) und nennt 99 Literaturangaben. Gefördert von: National Science Foundation.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Despite rising concerns about sycophancy—excessive agreement or flattery from artificial intelligence (AI) systems—little is known about its prevalence or consequences. We show that sycophancy is widespread and harmful. Across 11 state-of-the-art models, AI affirmed users’ actions 49% more often than humans, even when queries involved deception, illegality, or other harms. In three preregistered experiments ( N = 2405), even a single interaction with sycophantic AI reduced participants’ willingness to take responsibility and repair interpersonal conflicts, while increasing their conviction that they were right. Despite distorting judgment, sycophantic models were trusted and preferred. This creates perverse incentives for sycophancy to persist: The very feature that causes harm also drives engagement. Our findings underscore the need for design, evaluation, and accountability mechanisms to protect user well-being.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

ConsensusDeceptionIntentionSocial BehaviorLarge Language ModelsAdultFemaleHumansMaleYoung AdultMotivation

Autor:innen und Institutionen

  • Department of Computer Science, Stanford University, Stanford, CA, USA.
  • Department of Psychology, Stanford University, Stanford, CA, USA.
  • Human-Computer Interaction Institute, Carnegie Mellon University, Pittsburgh, PA, USA.

ORCID-Profile: Myra Cheng, Cinoo Lee, Pranav Khadpe, Dyllan Han, Dan Jurafsky

Förderung

  • National Science Foundation

Zitieren

APA

Cheng, M., Lee, C., Khadpe, P., Yu, S., Han, D., & Jurafsky, D. (2026). Sycophantic AI decreases prosocial intentions and promotes dependence. Science, 391(6792). https://doi.org/10.1126/science.aec8352

Vancouver

Cheng M, Lee C, Khadpe P, Yu S, Han D, Jurafsky D. Sycophantic AI decreases prosocial intentions and promotes dependence. Science. 2026;391(6792). doi:10.1126/science.aec8352

BibTeX
@article{Cheng2026,
  author = {Cheng, Myra and Lee, Cinoo and Khadpe, Pranav and Yu, Sunny and Han, Dyllan and Jurafsky, Dan},
  title = {Sycophantic AI decreases prosocial intentions and promotes dependence},
  journal = {Science},
  year = {2026},
  volume = {391},
  number = {6792},
  doi = {10.1126/science.aec8352}
}
Kostenlos starten →