Enabling Interoperable AI in IoT: A Unified Data Framework Approach
Aseem Khanna, Anuradha Bhusri · 2025
The integration of Artificial Intelligence (AI) into Internet of Things (IoT) systems is transforming how modern enterprises automate operations, make real-time decisions, and predict future outcomes. While many industries have adopted AI to enhance domains such as logistics, customer service, predictive maintenance, and supply chain optimization, most deployments remain isolated due to interoperability challenges. This paper investigates the structural limitations within existing IoT back-end architectures that prevent AI applications from interacting across functional boundaries. Using organizational surveys and a comprehensive review of related research, the study identifies key inhibitors, including non-standardized data flows, fragmented communication protocols, and incompatible device ecosystems. These factors constrain AI effectiveness by restricting cross-domain integration and collaboration. To address this, we propose the Unified IoT Systems AI Platform— a modular and interoperable architectural framework that supports secure, cross-functional AI deployment. The framework is designed to promote real-time orchestration, consistent data access, and secure AI model governance in heterogeneous enterprise environments.