Paper Type
Complete
Abstract
As artificial intelligence (AI) systems are increasingly used in high-stakes decision-making across organizations, the opacity of complex AI models poses significant challenges to user trust and model transparency thereby impeding their adoption. Explainable Artificial Intelligence (XAI) has emerged to address this gap, yet existing research mostly focuses on algorithmic and model-centric explanations with limited attention to how explanations shape transparency and trust from an Information Systems perspective. This study synthesizes extant XAI literatures to examine how various explanation methods contribute to transparency, understandability, and trust calibration in AI-driven decision-making. The findings reveal four distinct user-centric explanation paradigms - interpretive, cognitive, interactive, and temporal, that shape how users perceive and adjust transparency and trust in AI systems. Building on these insights, the study proposes a conceptual framework linking explanation paradigms to transparency and trust-related outcomes, and identifies key challenges and future directions for designing human-centric, interactive, and trustworthy XAI systems.
Paper Number
1367
Recommended Citation
Malla, Deependra and El-Gayar, Omar, "Enhancing Transparency and Trust in Explainable Artificial Intelligence: Insights and Opportunities" (2026). AMCIS 2026 Proceedings. 10.
https://aisel.aisnet.org/amcis2026/sig_dsa/sig_dsa/10
Enhancing Transparency and Trust in Explainable Artificial Intelligence: Insights and Opportunities
As artificial intelligence (AI) systems are increasingly used in high-stakes decision-making across organizations, the opacity of complex AI models poses significant challenges to user trust and model transparency thereby impeding their adoption. Explainable Artificial Intelligence (XAI) has emerged to address this gap, yet existing research mostly focuses on algorithmic and model-centric explanations with limited attention to how explanations shape transparency and trust from an Information Systems perspective. This study synthesizes extant XAI literatures to examine how various explanation methods contribute to transparency, understandability, and trust calibration in AI-driven decision-making. The findings reveal four distinct user-centric explanation paradigms - interpretive, cognitive, interactive, and temporal, that shape how users perceive and adjust transparency and trust in AI systems. Building on these insights, the study proposes a conceptual framework linking explanation paradigms to transparency and trust-related outcomes, and identifies key challenges and future directions for designing human-centric, interactive, and trustworthy XAI systems.
When commenting on articles, please be friendly, welcoming, respectful and abide by the AIS eLibrary Discussion Thread Code of Conduct posted here.
Comments
SIG DSA