Explainable Artificial Intelligence in Decision Support Systems: Origins and Applications

Explainability in Artificial Intelligence has been a concern since the rule-based expert systems of the 1970s and 1980s, which provided transparent decision rationales despite limited predictive power. With the advancement of statistical machine learning and deep learning in the 1990s and 2010s, model performance improved, but interpretability declined, leading to opaque “black box” systems whose internal logic is difficult to understand. This complexity increased the demand for transparency, particularly in high-stakes decision-making contexts. The launch of the DARPA XAI program in 2016 marked a turning point by formally recognizing the need for accurate yet interpretable models. Since then, Explainable Artificial Intelligence (XAI) has evolved as a key research area focused on generating human-understandable explanations that enhance trust, accountability, and usability. This chapter examines the origins, core concepts, and applications of XAI, emphasizing its role in supporting responsible AI deployment.