Explainable AI over the Internet of Things (IoT): Overview, State-of-the-Art and Future Directions

XAI builds trust in IoT devices by making AI decisions understandable to users. As billions of connected devices collect and act on data in our homes, workplaces, and cities, explainability helps end-users understand why these devices make specific decisions—a critical requirement for widespread adoption of IoT technology.

XAI is particularly important for high-stakes IoT applications like healthcare and industrial systems. The paper highlights Internet of Medical Things (IoMT) and Industrial IoT (IIoT) as key application areas where explainable AI can enhance security and help users trust automated decisions that directly affect safety and operations.

Edge computing enables XAI to work efficiently on IoT devices themselves. Rather than sending all data to distant servers, "edge XAI structures" allow AI explanations to be generated locally on devices, reducing delays and privacy risks—important for real-time applications like autonomous vehicles or medical monitoring.

XAI supports security in IoT networks by making threat detection transparent. When AI systems identify suspicious activity or potential cyberattacks in connected devices, explainability helps security teams understand why something was flagged as dangerous, enabling faster and more informed responses.

Future IoT systems will need XAI built in from the start, not added later. The paper emphasizes that as 6G networks and increasingly complex IoT ecosystems emerge, explainability must be integrated into the design of these systems rather than treated as an afterthought.