The convergence of AI and the IoT is reshaping data requirements, infrastructure needs and operational models, according to a white paper from Transforma Insights and Tata Communications (www.tatacommunications.com).

The report examines the need for what it calls an “intelligent last mile” to support the demands of AI in IoT.

AI and the IoT are two of the most impactful and far-reaching technology developments. There has been increasing discussion recently of the convergence of the two in concepts such as AIoT and physical AI. AI’s dependence on real-world data elevates the role of connected devices, while the rise of AIoT is accelerating the shift of intelligence towards the edge.

This creates increased pressure on connectivity to manage larger data volumes, stricter latency demands, and more complex regulatory and security environments. At the same time, organisations seeking to monetise AI are turning to IoT to create tangible services and improved efficiency, reinforcing the value of optimised connectivity.

The report “The Intelligent Last Mile: How networks must leverage AI to address the evolving needs of IoT” finds that IoT connectivity must evolve significantly to meet these requirements. Future options must be designed to be secure across all layers, compliant with fast-moving regulatory frameworks, and flexible enough to operate across global networks and multiple technologies.

They must also support deep interoperability across heterogeneous device fleets, orchestrate distributed AI workloads between edge and cloud environments, and enable collaboration across the IoT stack and between connectivity providers. In addition, platforms must become more user-friendly through unified management, automation and operational simplification. These capabilities collectively define the systems required for AI-enabled IoT.

The report examines a range of themes critical to how connectivity will need to evolve to address the needs of the convergence of AI and IoT. It starts by exploring the growing interdependence of AI and IoT, and the extent to which AI will be a driver of IoT. It then proceeds to examine the challenges of delivering data for AI and the functional requirements of what the report terms the “intelligent last mile”, including security, compliance, flexibility, interoperability, orchestration, collaboration, usability, resilience, scalability, efficiency, determinism and observability.

The report (transformainsights.com/research/reports/intelligent-last-mile) concludes with a worked example showing how these principles apply to connected cars.