The global telecommunications industry is undergoing a profound architectural shift. For decades, mobile network operators relied on massive, proprietary physical hardware appliances housed in local data centers to run their core operations. This traditional model, however, introduced high capital expenditures, slow service deployment times, and complex maintenance cycles.
To break this reliance, Nokia has announced an expanded, multi-year collaboration with Amazon Web Services (AWS) to deliver autonomous networks designed specifically for the artificial intelligence era.
Under this expanded partnership, Nokia will run its Autonomous Network Fabric directly on the AWS cloud platform. This integration allows telecommunications providers to migrate their entire operations and maintenance (O&M) stack to the public cloud, giving them immediate access to the advanced artificial intelligence and machine learning tools required to achieve Level 4 network autonomy.
Rather than relying on teams of human experts to manually oversee separate mobile, fixed, and transport networks, operators can now deploy an intelligently interconnected system that automatically manages itself at machine speed.
This technological pivot comes at a highly strategic moment for Nokia. The Finnish telecommunications giant reported a robust $23.1 billion in revenue over the last twelve months as of the first quarter of fiscal year 2026, carrying a solid gross profit margin of 45%.
Despite recent market momentum, the company’s stock currently trades near $6.04, which sits significantly below its independent fair value projection of $9.58.
By deepening its partnership with AWS, Nokia is reinforcing its position as a primary software and infrastructure leader, capitalizing on the massive AI supercycle to transform telecommunications from a static utility into a dynamic, programmable software platform.
The Core Framework: Understanding the Autonomous Network Fabric
To achieve Level 4 network autonomy—a state where the system can automatically monitor, analyze, and optimize itself across diverse operational environments without human intervention—Nokia has designed a comprehensive software architecture. The Autonomous Network Fabric integrates four core capabilities to streamline end-to-end workflows.
Unified Data Management Across Fragmented Silos
A major engineering challenge facing modern telecommunications providers is data fragmentation. A typical mobile network relies on hundreds of siloed operational and business support systems, each running on its own data architecture and proprietary databases. This fragmented environment makes it incredibly difficult to apply machine learning models consistently across different domains.
Nokia’s Autonomous Network Fabric solves this bottleneck by establishing a unified data management layer across all domains. This architecture is closely aligned with Nokia’s recent development work with Databricks.
By demonstrating a unified, substrate-agnostic data platform, the company proved that operators can simplify their complex data environments and ingest massive streams of real-time network data without needing to rewrite code.
This clean, unified data foundation serves as the essential raw material that feeds the underlying AI agents, allowing them to make highly coordinated, cross-domain decisions in real-time.
Agentic AI for Proactive Service Operations
Traditional network automation relies heavily on rigid, rule-based systems. While effective for simple, predictable tasks, these systems cannot keep pace with the extreme complexity of modern 5G-Advanced networks, which require thousands of predefined rules to manage every possible scenario permutation.
To overcome this limitation, Nokia is embedding “agentic AI” directly into its operational stack. Unlike traditional, reactive software, agentic AI utilizes advanced foundation models to reason, adapt, and cooperate with other specialized agents to achieve complex goals.
By integrating with AWS cloud services like Amazon Bedrock, these AI agents can analyze real-time performance data, identify subtle traffic anomalies, and coordinate with neighboring agents to optimize network configurations dynamically.
This shift to agentic AI allows the network to anticipate changing traffic mixes and adjust itself proactively, ensuring an optimal customer experience without requiring human intervention.
Digital Twin Simulations for Risk Mitigation
Operating a live telecommunications network is a high-stakes endeavor. If a network engineer applies a faulty configuration change to a busy base station, it can instantly trigger a localized blackout, disrupting services for thousands of businesses and emergency callers.
The Autonomous Network Fabric mitigates this risk by incorporating advanced digital twin simulations. Before applying any automated, AI-recommended optimization workflow to the live network, the system runs a proactive impact assessment inside a highly precise digital twin sandbox.
This virtual environment simulates the exact physical and logical parameters of the network, allowing the software to test the changes and verify their safety.
If the simulation identifies a potential conflict, the system automatically refines the recommendation, protecting the live network from accidental downtime and ensuring maximum operational stability.
The Definitive Prohibition of Manual Overrides
To ensure the safety of these closed-loop automated actions, the platform operates under a strict, intent-based networking model. Operators do not program individual commands or configure manual settings. Instead, they define their high-level business goals and service-level agreements (SLAs) through a simple, natural-language interface.
The system’s intent engine translates these business goals into automated, closed-loop actions, continuously monitoring the network to verify that the SLAs are being met.
If the system detects a potential violation—such as a sudden surge in data demand during a major sports event—it automatically deploys additional network slices, prioritizes critical traffic, and rebalances the workload.
By automating these complex, real-time adjustments, the platform ensures that the network continuously aligns itself with the operator’s business objectives, leaving human engineers to focus on high-level strategy and governance.
The Physical Implementation on AWS: Scalability and the Cloud Footprint
Running an exascale-class autonomous network platform requires a massive, highly scalable cloud infrastructure. By deploying the Autonomous Network Fabric on AWS, Nokia is leveraging the full power of the world’s leading cloud provider to deliver global availability and elastic scalability.
Leveraging Amazon Bedrock and SageMaker
The technical integration with AWS is designed to provide operators with direct access to advanced machine learning pipelines. The platform utilizes Amazon Bedrock to access a wide variety of high-performance foundation models, including Amazon Titan, Claude, and Llama, which serve as the cognitive engines for Nokia’s agentic AI.
Additionally, the system integrates with Amazon SageMaker to manage the end-to-end lifecycle of the custom machine learning models used for anomaly detection and root cause analysis.
By running these pipelines on AWS’s global, highly secure infrastructure, telecommunications providers can scale their operational ambitions instantly, deploying advanced AI capabilities across millions of connected devices without needing to purchase and maintain expensive on-premise computing hardware.
An Optimized, Cost-Efficient Cloud Footprint
While public cloud infrastructure offers unparalleled scalability, running constant AI workloads can quickly become prohibitively expensive if the software is not properly optimized. To address this financial bottleneck, Nokia has engineered an optimized cloud footprint for the Autonomous Network Fabric.
By utilizing containerized microservices and advanced cloud-native design principles, Nokia has succeeded in reducing the compute and storage requirements of the platform by up to 35% compared to traditional on-premises deployments.
The software leverages AWS’s elastically scalable compute instances, automatically spinning down unused processing resources during periods of low network activity and spinning them back up during peak hours.
This highly efficient resource utilization lowers the operator’s ongoing cloud bills and significantly reduces the carbon footprint associated with running complex network operations, helping telcos meet their environmental and sustainability targets.
The Practical Performance Metrics of Level 4 Autonomy
The transition to Level 4 autonomy is already delivering highly impressive, real-world financial and operational benefits to early adopters of Nokia’s autonomous networks portfolio.
Surpassing Ninety Percent Automation Rates
According to official performance data released by Nokia, telecommunications operators utilizing its autonomous software suite have achieved automation rates exceeding 90% across their core operations.
This high level of automation allows lean engineering teams to manage highly complex, multi-domain networks, reducing the administrative overhead associated with daily O&M tasks.
The speed of service delivery has also experienced a massive improvement. Traditional, manually configured service rollouts often require weeks of preparation, testing, and manual provisioning.
By using Nokia’s intent-based orchestration tools, operators can compress these timelines significantly, reducing service delivery times to four hours or less.
Additionally, the platform has cut the time required to roll out new network slices by up to 85%, allowing telcos to capitalize on new market opportunities and deliver on-demand connectivity to corporate clients within minutes.
Drastic Reductions in Outages and Customer Incidents
The closed-loop, automated remediation capabilities of the Autonomous Network Fabric are also delivering unparalleled levels of network reliability.
By continuously monitoring the system, identifying anomalies, and executing automated fixes in real-time, the platform has succeeded in reducing customer-impacting incidents by up to 50%.
When a system fault does occur, the software’s rapid, closed-loop resolution capabilities ensure that the impact is minimized.
Operators running Nokia’s autonomous portfolio have managed to limit total service interruptions to one minute per year or fewer.
This extreme level of reliability is critical for modern industries like smart manufacturing, connected health, and autonomous logistics, where even a brief network outage can result in massive financial losses or threaten public safety.
The Strategic Alliances: From MWC 2026 to Vodafone Trials
The expanded partnership between Nokia and AWS is not an isolated announcement; it is the culmination of a series of successful, high-profile joint activities that have demonstrated the viability of telco cloudification.
Agentic AI-Driven 5G Slicing at MWC 2026
Earlier in the year, during the Mobile World Congress (MWC) in Barcelona, Nokia and AWS captured the attention of the global tech community by showcasing the first live deployment of an agentic AI-driven 5G-Advanced network slicing solution.
The demonstration was conducted in the live, operating networks of major international carriers du and Orange.
Unlike traditional, static network slices that require manual reconfiguration to handle changes in demand, this breakthrough solution uses agentic AI running on Amazon Bedrock to automatically analyze real-time context data, including traffic volumes, geographic locations, and even local weather patterns.
By adjusting the network slices dynamically in response to real-world events, the solution proved that AI can deliver context-aware, premium connectivity on demand, providing a successful blueprint for the commercial deployment of Level 4 autonomous networks.
Vodafone’s IoT Voice and Data Core AWS Trial
The viability of running critical core network functions on public cloud infrastructure was further validated by a successful joint trial conducted with Vodafone.
During the wide-ranging trial, Vodafone, Nokia, and AWS successfully deployed a proof of concept of Nokia’s mobile packet core and IP Multimedia Subsystem (IMS) voice core on AWS cloud infrastructure in Frankfurt, Germany.
The trial successfully supported high-value, low-latency Internet of Things (IoT) services, including smart metering for utilities and emergency calling applications in vehicles and elevators.
By hosting these critical, cloud-native network functions on AWS Elastic Kubernetes Service (EKS) and Elastic Compute Cloud (EC2) instances, Vodafone demonstrated that tier-one operators can achieve the elasticity, scalability, and operational agility required to support over 240 million managed IoT connections globally.
This successful trial proved that the public cloud is fully ready to handle the most demanding, mission-critical workloads of the modern telecommunications industry, setting the stage for widespread commercial adoption.
A New Era of Autonomous Connectivity
The expanded collaboration between Nokia and Amazon Web Services marks a historic turning point for the telecommunications industry. By porting the Autonomous Network Fabric to the AWS platform and leveraging advanced AI services like Amazon Bedrock and SageMaker, the two companies are transforming telecommunications from a rigid, hardware-bound utility into a highly agile, software-defined, AI-native platform.
This transition is no longer a distant theoretical vision; it is a solid business imperative.
With operators already achieving automation rates exceeding 90%, reducing service delivery times to under four hours, and cutting network slice rollout times by 85%, the financial and operational benefits of telco cloudification are undeniable.
As we move deeper into the AI era, the collaboration between Nokia and AWS will continue to lead the way, helping telecommunications providers compress years of traditional network transformation into a few months of rapid cloud deployment, securing a highly efficient, resilient, and profitable future for global connectivity.
