This paper introduces Explainable Artificial Intelligence (XAI) as a requisite solution for increasing the interpretability and trustworthiness of ML systems. This paper explores the importance of XAI through case studies in two key sectors: manufacturing and healthcare. The first case relates to a predictive maintenance application that uses XAI to anticipate the likelihood of machinery failure using a gradient-boosting decision tree, which provides detailed recommendations for optimizing productivity. The second case study is based on the healthcare sector. LIME and SHAP tools for AI model explanation enhance trust in the effectiveness of diabetes predictions by AI-aided medical diagnosis. Both cases prove that XAI helps better understand the workings of exhibited machine learning models and aligns with ethical decision-making and regulatory requirements. Therefore, in the last part of the paper, the authors discuss some of the drawbacks of current forms of XAI and potential advancements in this field, including horizontality and better incorporation into the systems that use AI. This discovery supports using XAI to foster more transparent, responsible, and trustful AI applications in numerous industries.
