Transparency in AI: A Review of Explainable Artificial Intelligence Techniques

This article presents an overview of Explainable Artificial Intelligence (XAI) and its various methods, emphasizing the role of interpretability, in present day AI systems. With AI becoming more prevalent in decision making processes across fields there is a growing need for transparency and accountability. We delve into XAI techniques, such as agnostic methods like Local Interpretable Model agnostic Explanations (LIME) and SHapley Additive exPlanations (SHAP) model specific approaches like Integrated Gradients and transparent models like decision trees and rule based systems. By examining the strengths and limitations of these methods we provide insights into their usefulness and effectiveness, in scenarios. This review aims to give researchers and practitioners a nuanced understanding of XAI to support the creation and implementation of transparent AI systems.