As data-driven decision-making sen'ices are being infused into Internet of Things (IoT) applications, especially at the 5G networks, Artificial Intelligence (Ai) algorithms such as deep learning, reinforcement learning, etc. are being deployed as monolithic application services for autonomous decision processes based on data from MT devices. however, for latency sensitive loT applications such as health-monitoring or emergency-response applications, it is inefficient to transmit data to the Cloud data centers for storage and AI based processing. In this article, 5G integrated architecture for intelligent loT based on the concepts of AI cis a microservice (AIMS) is presented. The architecture has been conceived to support the design and development of Al microservices, which can be deployed on lederated and integrated 5G networks slices to provide autonomous units of intelligence at the Edge of Things, cis opposed to the current monolithic loT-Cloud services. The proposed 5G based Al system is envisioned as platfi-m for effective deployment of scalable, robust, and intelligent cross-border loT applications to provide improved quality of experience in scenarios where realtime processing, ultra-low latency and intelligence are key requirements. Finally, we highlight some challenges to give.future research directions.