Hide or Highlight: Understanding the Impact of Factuality Expression on User Trust
Open Access
- 15 October 2025
- journal article
- Published by Association for the Advancement of Artificial Intelligence (AAAI) in Proceedings of the AAAI/ACM Conference on AI, Ethics, and Society
- Vol. 8 (1) , 785-797
Abstract
Large language models are known to produce outputs that are plausible but factually incorrect. To prevent people from making erroneous decisions by blindly trusting AI, researchers have explored various ways of communicating factuality estimates in AI-generated outputs to end-users. However, little is known about whether revealing content estimated to be less factual influences users' trust when compared to hiding it altogether. We tested four different ways of disclosing an AI-generated output with factuality assessments: transparent (highlights less factual content), attention (highlights factual content), opaque (removes less factual content), ambiguity (makes less factual content vague), and compared them with a baseline response without factuality information. We conducted a human subjects research (N=148) using the strategies in question-answering scenarios. We found that the opaque and ambiguity strategies led to higher trust while maintaining perceived answer quality, compared to the other strategies. We discuss the efficacy of hiding presumably less factual content to build end-user trust.Keywords
All Related Versions
This publication has 18 references indexed in Scilit:
- Atomic Fact Decomposition Helps Attributed Question AnsweringIEEE Transactions on Knowledge and Data Engineering, 2025
- Detecting hallucinations in large language models using semantic entropyNature, 2024
- Designing Transparency for Effective Human-AI CollaborationInformation Systems Frontiers, 2022
- How to Evaluate Trust in AI-Assisted Decision Making? A Survey of Empirical MethodologiesProceedings of the ACM on Human-Computer Interaction, 2021
- Show or suppress? Managing input uncertainty in machine learning model explanationsArtificial Intelligence, 2021
- Application of Expectancy Violations Theory to communication with and judgments about embodied agents during a decision-making taskInternational Journal of Human-Computer Studies, 2016
- The impact of initial consumer trust on intentions to transact with a web site: a trust building modelThe Journal of Strategic Information Systems, 2002
- Natural language question answering: the view from hereNatural Language Engineering, 2001
- Ambiguity as strategy in organizational communicationCommunication Monographs, 1984
- EQUIVOCATION: CHARACTER INSURANCEHuman Communication Research, 1975