PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
September 30, 2025Open Access

Dialogues with Large Language Models reduce conspiracy beliefs even when the AI is perceived as human

View Full Paper
Ask AI
Bookmark
Share

Authors

EBEsther BoissinTCThomas H. CostelloDADaniel J. Alonso

Discussion

Loading...

Member takes

Overview

Experiment shows AI interactions lower conspiracy beliefs in 955 participants, indicating persuasive potential of AI.

Key Points

  • Conversations with AI reduced confidence in conspiracy beliefs significantly and effectively, challenging their resistance.
  • Participants reported lower confidence in both conspiracy and unwarranted beliefs after interacting with AI, underscoring AI’s persuasive capabilities.
  • Randomized conditions assessed whether the AI's characterization as a tool or human impacted effectiveness but showed no significant differences.
  • Results support the idea that compelling messages are crucial for persuasion, regardless of whether an AI or human delivers them.

Cite This Study

Boissin et al. (2025) studied this question.

synapsesocial.com/papers/68dc26188a7d58c25ebb25bdhttps://doi.org/10.31234/osf.io/apmb5_v4
View Full Paper
Ask AI
Bookmark
Share