ABSTRACT
Explainable artificial intelligence (XAI) is an artificial intelligence coded in such a way that the artificial intelligence model’s internal working mechanism, the purpose of its component design, logical reasoning, and decision-making process are described to the average user. XAI usually reveals the merits and demerits of an AI model, the AI model inputs and its criteria which lead to a certain output, the reasoning behind settling down to a particular decision as opposed to alternative decisions, the trust-ability of decisions, probability and reasoning behind errors and possible error correction. This chapter provides an overall introduction to the theory and concepts of XAI for the AI community. This chapter describes the working mechanism of XAI, the XAI’s technique to support the reasoning process and its superiority over human reasoning, applications and impact areas of XAI, the research challenges associated with XAI, and the limitations of the XAI systems. The use cases of XAI are also explained while focusing on the popular use case areas of social media and the healthcare system.
