Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Understanding XAI Through the Philosopher's Lens: A Historical Perspective

Martina Mattioli, Antonio Emanuele Cinà, Marcello Pelillo

TL;DR

The paper argues that explainable AI (XAI) can gain theoretical grounding by linking its debates to the long-standing philosophy of science. It draws parallels between the historical evolution of scientific explanation (from deductive to probabilistic, pragmatic, and mechanistic accounts) and the current XAI landscape (from transparent rule-based systems to opaque models with surrogate explanations). Through a structured review of philosophical models (DN, IS, SR, pragmatics, unification, abduction, mechanistic, counterfactual) and their epistemological implications, the authors propose a bridge to interpretability, understanding, and evaluation criteria for XAI. The work highlights that pragmatic factors, model surrogacy, and counterfactual reasoning have practical significance for building trustworthy, explainable AI systems. The proposed cross-disciplinary lens aims to inform future XAI research and practice by grounding explanations in established philosophical concepts.

Abstract

Despite explainable AI (XAI) has recently become a hot topic and several different approaches have been developed, there is still a widespread belief that it lacks a convincing unifying foundation. On the other hand, over the past centuries, the very concept of explanation has been the subject of extensive philosophical analysis in an attempt to address the fundamental question of "why" in the context of scientific law. However, this discussion has rarely been connected with XAI. This paper tries to fill in this gap and aims to explore the concept of explanation in AI through an epistemological lens. By comparing the historical development of both the philosophy of science and AI, an intriguing picture emerges. Specifically, we show that a gradual progression has independently occurred in both domains from logical-deductive to statistical models of explanation, thereby experiencing in both cases a paradigm shift from deterministic to nondeterministic and probabilistic causality. Interestingly, we also notice that similar concepts have independently emerged in both realms such as, for example, the relation between explanation and understanding and the importance of pragmatic factors. Our study aims to be the first step towards understanding the philosophical underpinnings of the notion of explanation in AI, and we hope that our findings will shed some fresh light on the elusive nature of XAI.

Understanding XAI Through the Philosopher's Lens: A Historical Perspective

TL;DR

The paper argues that explainable AI (XAI) can gain theoretical grounding by linking its debates to the long-standing philosophy of science. It draws parallels between the historical evolution of scientific explanation (from deductive to probabilistic, pragmatic, and mechanistic accounts) and the current XAI landscape (from transparent rule-based systems to opaque models with surrogate explanations). Through a structured review of philosophical models (DN, IS, SR, pragmatics, unification, abduction, mechanistic, counterfactual) and their epistemological implications, the authors propose a bridge to interpretability, understanding, and evaluation criteria for XAI. The work highlights that pragmatic factors, model surrogacy, and counterfactual reasoning have practical significance for building trustworthy, explainable AI systems. The proposed cross-disciplinary lens aims to inform future XAI research and practice by grounding explanations in established philosophical concepts.

Abstract

Despite explainable AI (XAI) has recently become a hot topic and several different approaches have been developed, there is still a widespread belief that it lacks a convincing unifying foundation. On the other hand, over the past centuries, the very concept of explanation has been the subject of extensive philosophical analysis in an attempt to address the fundamental question of "why" in the context of scientific law. However, this discussion has rarely been connected with XAI. This paper tries to fill in this gap and aims to explore the concept of explanation in AI through an epistemological lens. By comparing the historical development of both the philosophy of science and AI, an intriguing picture emerges. Specifically, we show that a gradual progression has independently occurred in both domains from logical-deductive to statistical models of explanation, thereby experiencing in both cases a paradigm shift from deterministic to nondeterministic and probabilistic causality. Interestingly, we also notice that similar concepts have independently emerged in both realms such as, for example, the relation between explanation and understanding and the importance of pragmatic factors. Our study aims to be the first step towards understanding the philosophical underpinnings of the notion of explanation in AI, and we hope that our findings will shed some fresh light on the elusive nature of XAI.
Paper Structure (25 sections, 1 figure)

This paper contains 25 sections, 1 figure.

Table of Contents

  1. Introduction
  2. Related Works
  3. Previous Philosophical Contributions.
  4. Explanation and Social Aspects.
  5. The Call for Clarification.
  6. Philosophical Roots of Explanation
  7. Short History of XAI
  8. Black Box-ness Insights from Science
  9. Explanation Terminology Basics
  10. The Historical Evolution of Scientific Explanation Debate
  11. Pre-Hempelian Era
  12. The Received View
  13. Post-Hempelian Era
  14. Statistical Relevance Model.
  15. The Pragmatics of Explanation.
  16. ...and 10 more sections

Figures (1)

  • Figure 1: Timeline of Scientific Explanation and XAI. The upper line represents the chronological development of the philosophical models, while the lower line illustrates the evolution of XAI. The middle line represents the general gradual change from deductive explanations to statistical ones. Analogies have been highlighted, as shown by the legend.