Self-Explainability in Self-Adaptive and Self-Organising Systems: Status and Research Directions
A new paper reviews the status and future research directions for self-explainability (SX) in complex AI systems. The authors define SX as a system's ability to explain its own decision-making, going beyond traditional Explainable AI. Their systematic literature review reveals that most SX approaches are still conceptual, with limited practical implementations and no standardized evaluation methods, indicating a significant research gap. AI
IMPACT Highlights the need for standardized evaluation and practical implementation of self-explaining AI systems, crucial for trust and understanding in complex AI applications.