Nokia and Amazon Web Services (AWS) introduced what they describe as the industry’s first agentic AI-powered 5G-Advanced network slicing solution operating in a live network. The companies combined Nokia’s slicing capabilities across RAN, transport and core with AWS AI services to create an intent-based system that dynamically adapts network resources based on real-world data inputs. UAE operator du and European carrier Orange are the first to explore the technology in their commercial environments.
The system uses agentic AI modules integrated with Amazon Bedrock to ingest and analyze open internet data such as traffic patterns, weather conditions, event schedules, maps, and incident reports, alongside operator network data. Based on inferred context, the platform automatically adjusts RAN policies and slicing parameters to meet defined service-level agreements. Nokia said the solution continuously monitors live KPIs including bitrate and latency and modifies policies in real time to address traffic surges, emergencies, or large-scale public events.
Nokia built the architecture on its 5G AirScale base stations, MantaRay SMO, Edge Slicing capabilities, and agentic AI modules, interfacing with Amazon Bedrock via APIs. AWS supports deployment using Amazon EKS Hybrid Nodes, allowing operators to run AI agents and network workloads across cloud and edge environments under unified Kubernetes management. Nokia will demonstrate the agentic AI slicing solution at Mobile World Congress 2026 in Barcelona.
- First live-network demonstration of agentic AI-powered 5G-Advanced slicing
- Combines Nokia AirScale RAN, MantaRay SMO, and Edge Slicing with Amazon Bedrock AI services
- Uses open internet data (events, traffic, weather, maps, incidents) plus operator data for contextual inference
- Dynamically adjusts RAN policies to meet enterprise SLAs and consumer QoS targets
- Initial operator participants: du (UAE) and Orange (Europe)
- Supports enterprise campuses, public safety, FWA, gaming, XR, smart cities, ports, and mass events
- Deployable across cloud and edge using Amazon EKS Hybrid Nodes
“This innovation marks a major milestone in the evolution of AI-native networks. By combining Nokia’s advanced network slicing capabilities with agentic AI, we are enabling operators to deliver premium, intent-based services that adapt dynamically to real-world conditions,” said Pallavi Mahajan, Chief Technology and AI Officer at Nokia.
Analysis: Nokia continues to advance its Autonomous Networks strategy by embedding AI into RAN and service orchestration, extending earlier work around intent-based networking and MantaRay SMO. The collaboration with AWS positions hyperscale cloud AI platforms as control-layer partners inside operator networks, aligning with similar AI-network initiatives from Ericsson, Huawei, and other cloud-native telecom infrastructure providers integrating AI-driven automation into mobile network operations.
