AI Network Slicing Moves into 5G-Advanced

In sider

28/08/2026

AI Network Slicing Moves into 5G-Advanced

Network slicing is moving into a more intelligent phase as 5G-Advanced adds new capabilities for managing increasingly demanding applications. Instead of treating a slice as a predefined configuration that remains largely unchanged, operators are working toward networks that can adapt resources to changing traffic, application requirements and service conditions.

At its core, AI network slicing allows multiple logical networks to operate over the same physical infrastructure, with each slice designed around particular performance requirements. GSMA identifies slicing as a core 5G capability for applications that need differentiated quality of service, including high bandwidth, low latency and predictable performance for areas such as industrial robotics and autonomous systems.

The arrival of 5G-Advanced strengthens the foundation for this model. 3GPP describes Release 18 as the first 5G-Advanced release, introducing further enhancements to the 5G system. Its work also includes enhancements around AI and machine learning for NG-RAN, including support for AI/ML-assisted network slicing and coverage and capacity optimisation.

This matters because future telecom infrastructure will have to support applications with very different requirements at the same time. A factory automation system may need highly predictable latency and reliability, while a large public event can suddenly generate a major increase in traffic. A single fixed network configuration cannot optimise equally for every situation.

AI can provide an additional intelligence layer by analysing network conditions and helping determine how resources should be allocated between slices.

From Virtual Networks to Intelligent Network Resources

Traditional network slicing already separates services according to defined requirements. The next step is making those slices more responsive.

An AI-enabled system can continuously examine information such as traffic levels, latency, network performance and service conditions. Instead of relying entirely on manual configuration, it can help determine when a slice needs additional capacity, when resources can be released and when network policies should change.

This is moving from predefined slicing toward adaptive slicing.

The development is already being demonstrated in live-network environments. In February 2026, an industry collaboration demonstrated an agentic AI-powered 5G-Advanced slicing solution in a live 5G network. The system monitored network KPIs including bitrate and latency, combined them with contextual information such as traffic, incidents, locations, events and weather, and automatically adjusted RAN policies to meet service requirements.

The demonstration covered the RAN, transport and core, showing why AI-driven slicing is becoming an infrastructure challenge rather than a feature confined to the radio network. It also used four operating modes: chatbot, on-demand, scheduled and autonomous.

That architecture points toward a broader change in telecom infrastructure. The network is not simply providing separate virtual paths for different services. It is beginning to acquire the intelligence needed to decide how those virtual resources should respond as conditions change.

Key takeaway: 5G-Advanced is adding the network capabilities needed to make slicing more adaptive, while AI provides the intelligence for responding to changing network and service conditions.

The shift is still developing. Network slicing itself is already part of 5G, while more advanced AI-driven and intent-based implementations are being demonstrated and commercialised progressively. The significance of AI network slicing is therefore not that slicing is new, but that the infrastructure is becoming capable of managing those slices with greater awareness of what the network and applications need in real time.

AI Network Slicing is Becoming More Adaptive

The role of AI network slicing is changing as 5G-Advanced moves toward more programmable and responsive network infrastructure. Traditional network slicing allows operators to create multiple logical networks across the same physical infrastructure, each with different performance requirements. The next step is to make those slices respond dynamically as traffic, application requirements and network conditions change. GSMA identifies network slicing as an important capability for AI applications that require differentiated bandwidth, latency and reliability.

This is where AI adds a new layer of intelligence. Instead of relying entirely on predefined policies, AI can analyse network telemetry and contextual information to help determine how resources should be allocated between slices. The objective is not simply to create more slices, but to make the infrastructure more responsive to what applications require at a particular moment.

A 2026 demonstration of agentic AI-powered 5G-Advanced slicing showed how this can work in a live 5G environment. The system monitored network indicators including bitrate and latency, combined them with information such as traffic, incidents, locations, maps and events, and automatically adjusted RAN policies to meet service requirements. The solution operated across the RAN, transport and core, rather than treating slicing as an isolated radio function.

That cross-domain capability is important. A network slice supporting an industrial application may need predictable performance, while a sudden event in a particular location can create a completely different demand profile. AI-driven infrastructure can potentially recognise those changes and adapt the relevant network resources rather than waiting for an operator to manually reconfigure the network.

Intent is Becoming the Interface Between Applications and the Network

The development of AI network slicing is also connected to intent-based networking. Instead of specifying every technical parameter manually, an operator or application can define the desired outcome, such as maintaining a particular latency or bandwidth level. The network’s intelligence and orchestration layers can then determine how to configure the underlying resources.

GSMA’s 2026 work on the mobile AI era links 5G-Advanced with AI agents, deterministic service requirements and increasingly automated network operations. It argues that emerging AI applications are placing greater demands on bandwidth, latency and reliability, increasing the need for networks that can adapt to those requirements.

The practical implications are already appearing in network trials. The 2026 live-network demonstration supported four operating modes: chatbot, on-demand, scheduled and autonomous. This shows a progression from an AI system assisting an operator toward infrastructure that can execute defined slicing actions automatically within the network.

The same demonstration also illustrates how slicing can incorporate real-world context. For mass events, for example, the system can use information about events and locations to anticipate periods of high demand and prepare network capacity accordingly. For emergencies, it can support on-demand slicing for public-safety requirements.

Key takeaway: Network slicing is evolving from a static virtualisation capability toward a more adaptive infrastructure layer that can respond to changing service and network conditions.

The change is therefore not about replacing network slicing with AI. It is about adding intelligence to the infrastructure that already performs the slicing, allowing 5G-Advanced networks to move closer to intent-driven and dynamically managed operation.

Conclusion

AI network slicing is becoming an important part of the evolution toward next-generation telecom infrastructure because it can make shared network resources more responsive to changing application and service requirements. 5G-Advanced provides the capabilities needed to support more programmable and automated networks, while AI adds the intelligence needed to adapt those resources dynamically.

The technology is still moving from demonstrations toward broader deployment. The emerging model is not simply more network slices, but intelligent slices that can be monitored, adjusted and orchestrated around specific performance requirements.

As telecom infrastructure becomes more software-driven, the combination of 5G-Advanced, network slicing, AI and intent-based orchestration could allow operators to build networks that respond more quickly to traffic, service conditions and application demands without relying on manual configuration for every change.