Reducing belief in conspiracy theories as they unfold using large language models

Source

• Source name: arXiv
• Source kind: paper
• Authors / organization: Thomas H. Costello, Nathaniel Rabb, Michael Nicholas Stagnaro, Gordon Pennycook, David Rand
• Venue: arXiv: cs.HC
• Published: 2026-08-06
• Collected: 2026-08-07
• URL: http://arxiv.org/abs/2608.06151v1
• PDF: https://arxiv.org/pdf/2608.06151v1
• DOI: UNKNOWN

Plain-Language Brief

Researchers tested if conversations with a large language model (LLM) can reduce belief in emerging conspiracy theories. They conducted experiments in the days following two significant events, and found that participants who engaged in multi-turn conversations with an LLM about reducing their conspiracy beliefs showed significantly lower conspiracy beliefs compared to control groups.

What the Document Says

• The study investigates whether interactions with a large language model can reduce belief in newly emerging conspiracy theories following major events.
• Experiments were conducted after the July 2024 assassination attempt on Donald Trump and the September 2025 assassination of Charlie Kirk, involving participants who held conspiratorial views about these events.
• Participants in the LLM treatment had conversations with the model aimed at reducing their conspiracy beliefs, while control groups either discussed an irrelevant topic with the LLM or viewed a static fact sheet.
• The results showed that participants in the LLM treatment had significantly lower conspiracy beliefs compared to control groups in both experiments.
• Evidence of downstream effects was observed as well, with reduced belief in different conspiracies one to two months later in response to subsequent crisis events.

Key Concepts Mentioned

• conspiracy theories
• large language model
• assassination attempts
• experiments
• belief reduction

Claims Made by the Document

• Conversations with a large language model can reduce belief in emerging conspiracy theories.
• Participants in the LLM treatment showed significantly lower conspiracy beliefs compared to control groups.

Evidence Used

• Results from experiments conducted after two significant events

Methods / Approach

• Conducting experiments involving multi-turn conversations with a large language model

Limitations Stated by the Authors

Not stated in source

Limitations Noted by the Agent

Agent interpretation — not from the source.

• The study does not provide information on the specific methodology used by the large language model to reduce conspiracy beliefs.
• It is unclear how these findings generalize to other types of misinformation or different demographic groups.

Open Questions

• What specific methodology was used by the large language model?
• How do these findings apply to other types of misinformation and demographic groups?

Notes for Future Q&A

May include interpretation. Labeled where it does.

The text lacks information on the specific approach taken by the large language model in reducing conspiracy beliefs. Additionally, it does not specify how these results might be applicable to various types of misinformation or different demographics.

Raw Abstract / Excerpt

The emergence of conspiracy theories in the wake of major events is a significant societal challenge. Here we test whether conversational dialogues with a large language model (LLM) can reduce belief in immediately unfolding conspiracies. In experiments conducted in the days following the July 2024 assassination attempt on Donald Trump and the September 2025 assassination of Charlie Kirk, U.S. adults (Experiment 1: N = 472; Experiment 2: N = 1035) holding conspiratorial views about the crisis event engaged in a multi-turn conversation with an LLM prompted to reduce their conspiracy belief. Compared to control participants who either discussed an irrelevant topic with an LLM or viewed a static fact sheet, participants in the LLM treatment showed significantly reduced conspiracy beliefs in both experiments. We also found evidence of downstream effects of the LLM treatment, observing reduced belief in different conspiracies one to two months later in the wake of subsequent crisis events. These results shed light on the psychology of emerging conspiracies and highlight the potential for scalable, cognitively-focused interventions to counteract misinformation in the immediate aftermath of high-profile societal events.