Beyond Explainable AI (XAI): An Overdue Paradigm Shift and Post-XAI Research Directions
This work addresses fundamental problems in XAI research, proposing a foundational shift for the broader AI/ML community to improve reliability and certification.
The study critiques Explainable AI (XAI) for deep neural networks and large language models, identifying empirical and conceptual limitations such as paradoxes and false assumptions, and proposes a paradigm shift toward reliable and certified AI development with four components including verification-focused Interactive AI and AI Epistemology.
This study provides a cross-disciplinary examination of Explainable Artificial Intelligence (XAI) approaches-focusing on deep neural networks (DNNs) and large language models (LLMs)-and identifies empirical and conceptual limitations in current XAI. We discuss critical symptoms that stem from deeper root causes (i.e., two paradoxes, two conceptual confusions, and five false assumptions). These fundamental problems within the current XAI research field reveal three insights: experimentally, XAI exhibits significant flaws; conceptually, it is paradoxical; and pragmatically, further attempts to reform the paradoxical XAI might exacerbate its confusion-demanding fundamental shifts and new research directions. To move beyond XAI's limitations, we propose a four-pronged synthesized paradigm shift toward reliable and certified AI development. These four components include: verification-focused Interactive AI (IAI) to establish scientific community protocols for certifying AI system performance rather than attempting post-hoc explanations, AI Epistemology for rigorous scientific foundations, User-Sensible AI to create context-aware systems tailored to specific user communities, and Model-Centered Interpretability for faithful technical analysis-together offering comprehensive post-XAI research directions.