THE ROLE OF EXPLAINABLE ARTIFICIAL INTELLIGENCE IN INCREASING TRUST IN AUTONOMOUS SYSTEMS

Abstract

The rapid development of autonomous systems across various sectors highlights the importance of building user trust, particularly given the "black box" nature of many Artificial Intelligence (AI) systems. Explainable Artificial Intelligence (XAI) has emerged as a crucial solution to address this challenge by improving the understandability, transparency, and accountability of AI models. This study presents a systematic literature review (SLR) to map publication trends, research methods, theories used, and application domains of XAI in the context of user trust in autonomous systems. By analyzing 30 relevant articles from the Scopus and Web of Science databases between 2020 and 2025, the study finds a consistent increase in academic interest in XAI and trust. Experimental/scenario-based methods and surveys are the most frequently used approaches, while Trust Theory and Technology Acceptance Model (TAM) are the dominant theoretical frameworks. The results show that XAI significantly improves user trust, especially when explanations are tailored to the user's context and characteristics. However, the effectiveness of XAI varies depending on the type of explanation and application domain. This study fills this literature gap by providing a comprehensive mapping and highlighting the mechanisms by which XAI can enhance trust, offering practical guidance for autonomous system developers. Limitations of the study include the database coverage and time period. Future research is recommended to conduct longitudinal analysis, multi-domain empirical testing, and integrate user psychological factors for a more holistic understanding and development of optimal XAI designs.