Explainable AI does not provide the explanations end-users are asking for
Authors:
- Savio RozarioSRSavio Rozario
- George ČevoraGČGeorge Čevora
Preprint
Abstract
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.Keywords
All Related Versions
This publication has 6 references indexed in Scilit:
- Large pre-trained language models contain human-like biases of what is right and wrong to doNature Machine Intelligence, 2022
- An algorithmic approach to reducing unexplained pain disparities in underserved populationsNature Medicine, 2021
- Weight Poisoning Attacks on Pretrained Models2020
- Explanation in artificial intelligence: Insights from the social sciencesArtificial Intelligence, 2019
- Artificial intelligence faces reproducibility crisisScience, 2018
- “Why Should I Trust You?”: Explaining the Predictions of Any Classifier2016