arXiv:2603.03585v2 Announce Type: replace Abstract: Misinformation is a growing societal threat, and susceptibility to misinformative claims varies across demographic groups due to differences in underlying beliefs. As Large Language Models (LLMs) are increasingly used to simulate human behaviors, we investigate whether they can simulate demographic misinformation susceptibility, treating beliefs as a primary driving factor. We introduce BeliefSim, a simulation framework that constructs demographic belief profiles using psychology-informed misinformation taxonomies and survey priors. We study
The proliferation of misinformation and the increasing sophistication of LLMs make simulating human susceptibility a timely and crucial research area.
This research provides a framework for understanding and potentially mitigating the spread of misinformation by simulating its demographic impact and driving factors, which is critical for societal stability and information integrity.
The ability to model demographic susceptibility to misinformation using LLMs introduces a new tool for understanding and combating online falsehoods, potentially enabling more targeted interventions.
- · AI researchers
- · Social scientists
- · Public health organizations
- · Governments
- · Misinformation purveyors
- · Social media platforms (if forced to implement findings)
- · Propagandists
LLMs can be effectively used to simulate complex human social phenomena, specifically demographic susceptibility to misinformation.
Understanding demographic belief profiles regarding misinformation allows for the development of more effective and targeted counter-misinformation strategies.
These simulation capabilities could eventually lead to AI-powered early warning systems for emergent misinformation campaigns, or even proactive educational interventions tailored to specific demographic groups.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.CL