Home // AIHealth 2025, The Second International Conference on AI-Health // View article

Determinants of User Trust in an AI-enabled System in the Development Stage

Authors:
Pi-Yang Weng

Keywords: XAI, interpretability, explainability, satisfaction, trust.

Abstract:
Explainable Artificial Intelligence (XAI) has pro- vided a noticeable foundation for user trust building in recent years, especially in the high-risk decision scenarios, such as medical and healthcare domains. Building trust in an AI-enabled system is one of the important issues for users, which would start from the development stage. User trust could be enhanced by understanding the so-called black-box model. However, trust could be built by an emotional factor like user satisfaction in addition to scientific factors, such as XAI. In this paper, we present a framework named Three-Pillar User Trust to identify the underlying determinants of user trust in an AI-enabled system. We propose that the introduction of XAI can enhance user trust in the stages of model evaluation and validation by improving their comprehensibility with the AI system outputs and algorithms. Moreover, we propose that user satisfaction, as an emotional factor, would be an important component to influence user trust. To validate our framework, we will recruit some students from one university to participate in our experiment. This research will aim to build a three-pillar user trust framework with model interpretability, user satisfaction, and instance explainability.

Pages: 68 to 71

Copyright: Copyright (c) IARIA, 2025

Publication date: March 9, 2025

Published in: conference

ISBN: 978-1-68558-247-0

Location: Lisbon, Portugal

Dates: from March 9, 2025 to March 13, 2025