Nokia and AWS Pilot Agentic AI to Automate Real-Time 5G Network Slicing

Nokia and AWS Pilot Agentic AI to Automate Real-Time 5G Network Slicing

In a landmark development for telecommunications infrastructure, Nokia and Amazon Web Services (AWS) have unveiled a groundbreaking agentic AI-powered solution for real-time 5G network slicing. Announced earlier this month, the pilot program marks a pivotal shift from static, pre-configured network slices to dynamic, AI-driven resource allocation that adapts instantaneously to traffic demands, latency requirements, and service-level agreements. According to The Fast Mode, the system leverages autonomous AI agents that continuously monitor network conditions and make operational decisions without human intervention—ushering in what industry experts are calling the era of "self-optimizing networks."

Unlike traditional network slicing, which requires manual provisioning and periodic adjustments, the Nokia-AWS solution employs machine learning models trained on petabytes of real-world network telemetry data hosted on AWS’s cloud infrastructure. These AI agents, referred to as "agentic" due to their goal-oriented, proactive behavior, can predict congestion, reroute traffic, and allocate bandwidth across slices dedicated to critical applications such as autonomous vehicle communication, remote surgery, and industrial IoT—all within milliseconds. The system was demonstrated at MWC 2026 as part of Nokia’s 5G-Advanced showcase, with IEEE ComSoc Technology Blog noting that the demonstration achieved a 68% reduction in slice reconfiguration latency compared to conventional methods.

Two major telecom operators have already onboarded the pilot: Du, the UAE-based telecommunications provider, and Orange, one of Europe’s largest carriers. As reported by Developing Telecoms, both operators are testing the system in live urban and industrial environments. Du is evaluating its use in smart city applications, including real-time traffic management and public safety communications, while Orange is focusing on enterprise-grade private 5G networks for manufacturing and logistics hubs. Early results indicate a 40% improvement in service reliability and a 35% drop in operational costs associated with manual network tuning.

The integration of AWS’s cloud-native technologies—including Amazon SageMaker for model training, AWS IoT Greengrass for edge computing, and Amazon EventBridge for event-driven automation—is central to the solution’s scalability. Nokia’s Cloud-Native Network Functions (CNFs) interface seamlessly with AWS’s infrastructure, enabling end-to-end orchestration across hybrid and multi-cloud environments. This architecture allows operators to deploy and manage thousands of network slices simultaneously, each with unique performance parameters, without overwhelming network operations teams.

Industry analysts warn that while the technology promises transformative gains, it also introduces new challenges. Autonomous decision-making raises questions about accountability, cybersecurity, and regulatory oversight. "If an AI reallocates bandwidth during a critical emergency call and causes a failure, who is liable?" asked Dr. Elena Márquez, a telecom policy expert at the International Telecommunication Union. "We need new governance frameworks before this scales globally."

Nonetheless, the Nokia-AWS collaboration is being hailed as a blueprint for the future of 5G. With 5G-Advanced standards nearing ratification and 6G research accelerating, this agentic AI approach could become the industry standard for network automation. As more operators join the pilot, the potential for AI-driven networks to unlock new revenue streams—from ultra-reliable low-latency services to on-demand slice marketplaces—grows ever more tangible.