What type of study is this?

This is a Pre-Registered Trial study.

September 12, 2025

AI reduces conspiracy beliefs even when presented as a human expert

View Full Paper

Ask AI

Bookmark

Authors

Preregistered experiment shows AI reduces conspiracy beliefs in 955 participants, indicating effectiveness transcends messenger identity.

AI significantly reduced participants' confidence in conspiracy beliefs and unwarranted beliefs, revealing its persuasive capability.
The debunking messages were effective regardless of whether the messenger was framed as an AI model or a human expert.
The study involved a preregistered experiment with 955 participants, focusing on beliefs prone to conspiratorial thinking.
Findings suggest that compelling messaging plays a critical role in persuasion, regardless of the perceived identity of the messenger.

Boissin et al. (2025) studied this question.

View Full Paper

Ask AI

Bookmark

September 12, 2025

View Full Paper

Ask AI

Bookmark

Authors

Preregistered experiment shows AI reduces conspiracy beliefs in 955 participants, indicating effectiveness transcends messenger identity.

AI significantly reduced participants' confidence in conspiracy beliefs and unwarranted beliefs, revealing its persuasive capability.
The debunking messages were effective regardless of whether the messenger was framed as an AI model or a human expert.
The study involved a preregistered experiment with 955 participants, focusing on beliefs prone to conspiratorial thinking.
Findings suggest that compelling messaging plays a critical role in persuasion, regardless of the perceived identity of the messenger.

Boissin et al. (2025) studied this question.

View Full Paper

Ask AI

Bookmark