Large language models can effectively convince people to believe conspiracies

Kari Jaaskelainen

10 Jan 2026 — 1 min read

Can AI talk you into a conspiracy?

In three preregistered experiments with 2,724 Americans, people chatted with GPT-4o about a conspiracy they felt unsure about. The AI was told either to debunk it or to argue for it.

With guardrails removed, the AI was just as good at increasing belief as decreasing it.
Standard GPT-4o showed similar effects—safety filters did little to stop bunking.
Pro-conspiracy AI was rated more positively and even boosted trust in AI.
A follow-up corrective chat reversed the newly formed conspiracy beliefs.
Simply instructing the model to use only accurate information sharply reduced its ability to promote conspiracies.

Bottom line: today’s LLMs can persuasively promote both truth and falsehood. But simple prompts and corrective interactions show promise for reducing harm.

Paper by Costello, Pelrine, Kowal, Arechar, Godbout, Gleave, Rand, and Pennycook (cs.AI, econ.GN). Read more: https://arxiv.org/abs/2601.05050v1

Paper: https://arxiv.org/abs/2601.05050v1

Register: https://www.AiFeta.com

#AI #LLMs #Misinformation #ConspiracyTheories #AIEthics #AISafety #GPT4o #SocialImpact #Research #BehavioralScience

Meet GRACE: a moral governor for safer, more transparent AI

AI agents are getting powerful—so how do we make sure they do the right thing, not just the effective thing? Meet GRACE, a reason-based moral governor that keeps AI behavior aligned with human norms by separating moral reasoning from goal-driven decision-making. * Moral Module: uses deontic logic and explicit reasons

Why some fine-tuned LLMs miss phishing—and how to fix it

Not all fine-tuned LLMs spot phishing equally. A new study tests Llama 3.1 8B, Gemma 2 9B, and Mistral on high-stakes phishing detection—and uses SHAP and mechanistic interpretability to reveal why models do (or don’t) generalize. * Architecture × data diversity matters: Gemma 2 9B hits state-of-the-art performance (F1

AI that explains itself—by following the science

AI that explains itself—by following the science Black-box AI can be powerful, but it often can’t tell us why it made a decision. Concept Bottleneck Models (CBMs) try to fix that by predicting human-understandable concepts first, then the final answer. The catch: standard CBMs ignore domain-specific cause-and-effect and

Meet GenomAgent: A Team of AI Specialists for Smarter Genomics Q&A

TL;DR: Finding reliable answers in genomics is hard. GenomAgent turns one big AI into a coordinating team of specialists—and beats the current leader by 12% on a key benchmark. Genomic facts live across many databases. Standard chatbots struggle because they can’t flexibly query those sources. GeneGPT added