5G SA Updates in 72 Countries: The Network Upgrade That Decides Who Gets 6G First

The State of the Rollout: 47 Countries, but Not All Equal

The GSA reported in its January 2026 assessment that at least 85 operators across 47 countries have launched 5G SA networks. That sounds like meaningful progress, and it is, but context matters. According to the GSA's broader tally, 383 operators across 146 countries have committed to 5G in some form, whether through tests, pilots, or live deployments. Of those, the 85 with genuine SA deployments represent less than a quarter.

Ericsson's own network data suggests roughly 360 5G networks are live worldwide, with global population coverage expected to reach approximately 60 percent by the end of 2025. But population coverage for basic 5G connectivity tells a different story than coverage for 5G SA, which requires a full core network upgrade, not just new radio hardware. The two figures are measuring fundamentally different things.

Think of it this way: NSA 5G is like installing a faster engine in an old car with the original transmission. The speed improvements are real, but the drivetrain limits how much power actually reaches the wheels. SA 5G replaces the whole drivetrain, enabling features, specifically network slicing, ultra-low latency, and AI-driven resource management, that were never possible on the older architecture.

Ookla's 2026 analysis of 5G SA reach shows that six of the top ten countries by 5G SA sample share are in Asia Pacific. China leads at 79.0 percent 5G SA sample share, India sits at 49.2 percent, and Singapore rounds out the front of the regional pack. The GCC states, the Nordic countries, and the United States are the other regions where SA deployments are the most advanced. Much of Europe, Latin America, and sub-Saharan Africa remains on the NSA path.

5G-Advanced: The Next Gate, and Who Has Passed It

5G-Advanced, formally defined in 3GPP Release 18, is the upgrade that sits between current 5G SA and eventual 6G. It brings deterministic latency, enhanced uplink performance, superior positioning accuracy, and, most critically, dynamic network slicing, the ability to allocate network resources in real time to specific applications or customers rather than carving out static slices in advance.

"We've identified about a dozen MNOs that have launched 5G-Advanced. The operators in China, China Mobile, China Telecom, China Unicom, were early adopters of 5G SA and have launched 5G-Advanced. In the UAE, Du and e& have also launched 5G-Advanced, and so has T-Mobile in the U.S."

Siân Morgan, Research Director, Dell'Oro Group

That is a short list. About a dozen operators, globally, out of roughly 360 with any 5G presence, have crossed into 5G-Advanced territory. Dell'Oro's Morgan notes that another 20 to 30 are actively working toward deployment, which means the near-term pipeline is real. But 30 to 40 out of 360 still leaves the vast majority of global operators running 5G capabilities that are frozen at the standard's first generation.

Recon Analytics analyst Daryl Schoolar is blunter about the pace of progress.

"The ecosystem needed for monetization is not there and is limited due to the low number of commercial deployments. At the rate of deployment, it could be 2028 before we near 50 percent of all commercial 5G networks that are also 5G-Advanced."

Daryl Schoolar, Analyst, Recon Analytics

The word "monetization" is the key one in that quote. Operators do not upgrade core networks purely on principle. They do it when there is a business case that justifies the capital expenditure. And right now, the enterprise and consumer use cases that would make 5G-Advanced financially compelling, namely dynamic slicing sold to enterprises for guaranteed quality of service on specific applications, are still nascent.

Why Dynamic Slicing Is the Whole Point

Network slicing, at its simplest, means dividing a single physical network into multiple virtual networks, each with guaranteed performance characteristics. A hospital's patient monitoring traffic gets its own slice with guaranteed low latency. A factory floor's autonomous robot traffic gets a separate slice with guaranteed reliability. A consumer streaming service gets another. Each runs on the same radio hardware but is logically isolated from the others.

The "dynamic" part is what 5G SA enables that older approaches could not. Static slicing, available to some degree on NSA networks, pre-allocates resources to each slice at setup time. If demand on one slice spikes unexpectedly, it cannot borrow capacity from an underused slice. Dynamic slicing, by contrast, can reallocate resources in milliseconds based on real-time demand across the whole network. A conference that suddenly generates a burst of video uploads does not degrade the emergency services slice running on the same infrastructure.

Verizon has already begun selling what it calls SLA-backed services using network slicing for business customers. T-Mobile has launched what it markets as SuperMobile, described as the first nationwide enterprise slicing product in the United States. These are early commercial examples, but they represent a fundamentally different way of monetizing wireless infrastructure: not selling gigabytes, but selling guaranteed performance.

Dell'Oro's Morgan notes that the AI and machine learning enhancements built into 5G-Advanced "will be a foundation for 6G," a point Ericsson echoed at Mobile World Congress in early 2026, suggesting that AI-driven core updates could arrive far faster than traditional 3GPP release cycles would imply.

The 6G Horizon and the Architecture Dependency

SoftBank's move is instructive precisely because it frames SA not as an endpoint but as a prerequisite. The carrier's stated goal is 6G services in 2029 to 2030, and its network modernization agreement with Ericsson is explicitly positioned as the foundation that makes that timeline achievable. Operators who have not yet migrated to SA will need to complete that migration before 6G deployment becomes technically feasible: the industry consensus is that 6G will be deployed on the 5G SA core, not on a new parallel core architecture.

That architecture dependency creates a compounding schedule problem. An operator that starts its SA migration in 2027 and takes 18 to 24 months to complete it across a national network will not be in position to deploy 6G until the early 2030s at the earliest. Operators that completed SA migrations in 2023 and 2024, the Chinese carriers and the UAE operators being the clearest examples, are already running 5G-Advanced features today and will be in line for 6G trials before many of their peers have finished basic SA buildouts.

What This Means for Enterprises and Consumers

For enterprise customers, the practical implication is straightforward: only operators running 5G SA can offer genuinely differentiated network slicing products. If a manufacturing company wants guaranteed sub-10 millisecond latency for its factory automation, it needs to be in a country where the local operator has made the SA transition and deployed 5G-Advanced slice management. That list of countries is still short.

For consumers, the near-term impact is more limited. The 5G speeds that matter most for consumer use cases, fast downloads, responsive streaming, reliable indoor coverage, are largely available on NSA networks. The difference is in the ceiling: SA networks have a much higher ceiling for throughput, latency consistency, and the capacity to handle dense device environments. In a stadium filled with 80,000 people all streaming video simultaneously, the difference between SA and NSA will be perceptible. In a suburban living room, it largely is not.

The more consequential consumer benefit of 5G-Advanced is the RedCap standard, which enables low-power 5G connectivity for wearables, health monitoring devices, and smart home sensors. Apple and several major operators are actively developing 5G RedCap smartwatch support, as reported by Fierce Network in early 2026. This expands the ecosystem of devices that can run natively on 5G infrastructure without draining batteries in hours.

The Operator Gap Is a Strategic Risk

The bifurcation in the 5G market has a parallel in the 4G era that is worth recalling. When LTE first deployed, there was a substantial period during which operators in advanced markets offered fundamentally different mobile internet experiences than those in less advanced markets. The gap closed eventually, but it took a decade, and it reshaped which device makers, app developers, and enterprise technology providers built for which markets first.

5G's architectural gap has the potential to be stickier, because the SA migration requires a full core network replacement rather than a radio upgrade. Operators that delayed the 4G-to-5G radio upgrade could catch up relatively quickly by deploying new base stations. Operators that need to replace their entire packet core are facing a multi-year project that competes for capex against every other priority on their balance sheet.

The GSA's figure of 85 SA operators in 47 countries will grow. Dell'Oro expects most operators that have launched SA to proceed to 5G-Advanced over the coming years. But Recon Analytics' 2028 projection for reaching 50 percent 5G-Advanced penetration among commercial 5G networks is a reminder that the technology industry's tendency to treat a standard's launch date as synonymous with broad availability is a persistent source of misaligned expectations.

The operators running 5G-Advanced today are not simply ahead in the deployment race. They are the ones writing the monetization playbook: learning which enterprise use cases pay, which slicing configurations customers actually want, and how dynamic resource management behaves under real-world conditions. That operational knowledge will compound over the next three to four years while the rest of the industry is still completing SA migrations. Whether that gap closes before 6G arrives, or widens further, is the central question for the wireless industry through 2030.