Explainable AI does not provide the explanations end-users are asking for
This addresses a critical gap in AI deployment for organizations and users seeking reliable explanations, though it is incremental as it critiques existing approaches rather than introducing new methods.
The paper argues that current Explainable AI (XAI) techniques fail to meet end-user needs for understanding and trust in AI systems, concluding that transparency and rigorous validation are more effective for building trust.
Explainable Artificial Intelligence (XAI) techniques are frequently required by users in many AI systems with the goal of understanding complex models, their associated predictions, and gaining trust. While suitable for some specific tasks during development, their adoption by organisations to enhance trust in machine learning systems has unintended consequences. In this paper we discuss XAI's limitations in deployment and conclude that transparency alongside with rigorous validation are better suited to gaining trust in AI systems.