Telenor unveils AI Slice - redefining the telco role in the AI economy
Published
Telenor unveils AI Slice – redefining the telco role in the AI economy
At this years Mobile World Congress (MWC) in Barcelona, Telenor unveils the AI Slice concept, developed through close collaboration across network, cloud and AI domains, and in partnership with leading technology innovators.
The AI Slice introduces a new architectural approach designed for the next wave of enterprise AI adoption. As AI workloads become more distributed, latency-sensitive and compute-intensive, businesses require more than traditional connectivity. They need seamless integration of the Telco environment where compute capacity and AI inference is one converged service, orchestrated across edge and cloud environments trough connectivity
AI Slice represents Telenor’s response to that need.
From connectivity provider to infrastructure orchestrator
Telecom networks have long connected enterprises to public clouds. With AI Slice, Telenor demonstrates how the telco can take a broader role: orchestrating infrastructure and connectivity as a unified capability.
AI workloads are no longer confined to centralized data centers. Many B2B use cases, such as industrial automation, video analytics, robotics, autonomous systems, mission-critical operations, demand compute resources closer to where data is generated. Performance, latency and data sovereignty increasingly matter as much as raw compute power.
The AI Slice concept builds on Telenor’s ongoing Proof of Concept in AI RAN, where teams have explored how cloud-native architectures and elastic GPU orchestration can support both:
- AI for RAN - real-time network intelligence and optimisation
- AI on RAN - enterprise AI workloads running across telco infrastructure
Through this work, the scope evolved beyond AI RAN into a broader architectural concept.
AI Slice integrates telco edge, core cloud and tenancy from Telenor AI Factory into a distributed telco cloud continuum. An elastic GPU orchestration framework aggregates resources across cloud domains, enabling AI workloads to be dynamically placed and prioritised based on performance requirements, latency sensitivity and resource availability - while safeguarding mission-critical network intelligence.
In practice, this means enterprises can accelerate AI innovation across distributed sites without managing fragmented infrastructure. Connectivity becomes an active component of the AI delivery chain – not just the transport layer.
A strategic concept for B2B transformation
For enterprise customers, the AI Slice demonstrates how telecom infrastructure can:
- Deliver AI compute closer to operations
- Support hybrid and multi-cloud strategies
- Combine network intelligence and enterprise AI workloads
- Provide a seamless experience across connectivity and compute
- Ensure performance of low latency use cases
Enterprises are already connected. With AI Slice, Telenor extends that connection into distributed AI infrastructure - strengthening the link between networks, clouds and edge computing environments.
This evolution positions the Telco as a neutral orchestrator across cloud domains, capable of integrating hyperscaler environments, telco edge infrastructure and AI factory capabilities into a cohesive operational framework.
Ecosystem co-creation
The AI Slice concept has been developed in collaboration with the key ecosystem partners Nvidia, Supermicro, Avesha and Vaidio, reflecting a shared ambition to bridge telecom and AI domains.
Together, the partners contribute expertise across GPU-optimized infrastructure, cloud-native networking, AI orchestration and enterprise AI applications.
“The AI economy requires a new approach to infrastructure,” said Karine Storaker Braaten, VP Innovation, Telenor. “Connectivity remains our foundation - but in the AI era, connectivity combined with distributed compute becomes a strategic enabler. Through ecosystem collaboration, we are exploring how telcos can help deliver AI capabilities across clouds and closer to where value is created.”
The results will be showcased at Mobile World Congress 2026.
