Author: Alan Weissberger
Ericsson and MediaTek Demonstrate 3GPP-Based GNSS RTK Positioning with Sub-30cm Accuracy Over a Commercial 5G Network
Ericsson and MediaTek have completed a world-first end-to-end demonstration of high-precision outdoor positioning over a commercial 5G network, using standardized 3GPP Global Navigation Satellite Systems (GNSS) Real-Time Kinematic (RTK) assistance. The trial achieved stable outdoor positioning accuracy below 30 cm using a handset’s integrated antenna, demonstrating a scalable path to decimeter-level positioning without proprietary positioning infrastructure or dedicated external GNSS hardware.
The demonstration validates how 5G networks can distribute GNSS RTK correction information efficiently to devices for industrial automation, autonomous mobility, drones, robotics, and other applications requiring reliable, mission-critical location awareness.
Image courtesy of Ericsson
………………………………………………………………………………………………………………………………………………………………………………………………………………………
Technical highlights:
-
Sub-30 cm positioning performance: The device achieved stable outdoor accuracy below 30 cm with its built-in antenna. With a geodetic-grade GNSS antenna, the same RTK approach can support centimeter-class positioning performance.
-
3GPP-standardized architecture: The solution uses capabilities specified in 3GPP Release 16 for GNSS positioning assistance, enabling an interoperable architecture across compatible networks, positioning servers, chipsets, and devices.
-
End-to-end unicast and broadcast validation: Both delivery modes were demonstrated end to end:
-
Unicast GNSS RTK: LPP assistance messages are delivered through Secure User Plane Location (SUPL) over standard IP connectivity, enabling targeted delivery to individual UEs.
-
Broadcast GNSS RTK: RTK assistance data is transmitted through Positioning System Information Blocks (PosSIBs), allowing common assistance information to be delivered simultaneously to all capable devices within a cell.
-
-
Efficient scaling model: Broadcast delivery can reduce repeated transmission of identical RTK correction data in high-density areas, while unicast can be used for lower-traffic scenarios, individualized service handling, or LTE-connected devices. This supports flexible selection of the most efficient assistance-delivery method based on coverage, device population, and radio-resource conditions.
-
Secured broadcast operation: While broadcast information is radio-accessible to devices in the coverage area, 3GPP mechanisms support encrypted positioning assistance. Authorized subscribers obtain the required decryption material through the 5G Core, including the Access and Mobility Management Function (AMF), protecting access to premium high-precision positioning services.
GNSS correction support:
The trial supports two standardized GNSS correction-data models:
-
Observation State Representation (OSR): Provides corrections derived for a specific receiver or location context, including GNSS observation-related corrections.
-
State Space Representation (SSR): Delivers correction parameters associated with GNSS satellite orbit, clock, bias, and atmospheric error states. SSR is particularly suitable for scalable service delivery because the common state information can be applied by many devices across a service area.
In both cases, the UE applies the received RTK assistance in real time to mitigate satellite, orbital, clock, ionospheric, tropospheric, and other GNSS error sources that limit conventional standalone GNSS accuracy.
Network and device implementation:
The test used Ericsson’s 5G Core, Ericsson Network Location (ENL), Ericsson 5G Advanced Location Services, and Ericsson RAN to generate, manage, and distribute OSR and SSR correction information through the mobile network. MediaTek validated the device side using its latest 5G modem technology with integrated GNSS capability, including real-time processing of advanced GNSS correction data with power-optimized implementation.
“This is a milestone for the 5G ecosystem,” said Johan Hultell, Head of Product RAN Software at Ericsson. “Together with MediaTek, we make high-precision location more accessible to device makers and enterprises, accelerating use in manufacturing, transport and beyond.”
Dr. HC Hwang, General Manager of Wireless Communication Systems and Partnerships at MediaTek, added: “The collaboration with Ericsson demonstrates how 5G Advanced technology can unlock new value for consumers and enterprises alike. Our modems with integrated GNSS are ready to support these advanced positioning services, paving the way for the next generation of intelligent devices.”
Technology backgrounder:
Conventional standalone GNSS typically provides positioning accuracy in the range of several meters. RTK improves this performance by using measurements from accurately surveyed reference stations to generate correction assistance. In the 3GPP architecture, this assistance is delivered through the cellular network using LPP-based positioning procedures and, where applicable, broadcast system information.
By standardizing the delivery of GNSS RTK assistance over 4G and 5G networks, operators can offer high-precision location capabilities at network scale. This creates a standards-based foundation for digitalized industrial operations, autonomous systems, intelligent transportation, asset tracking, and spatially aware consumer devices.
Accuracy to within tens of centimeters is a significant improvement compared to traditional GNSS, which is accurate to within several meters, and therefore paves the way for more advanced use cases. These include the safe operation of autonomous vehicles, drones and industrial robots in complex outdoor environments, Ericsson said. Complex environments encompasses places like factories, logistics hubs, ports, and mines etc.
…………………………………………………………………………………………………………………………………………………………………………………………………………………………………………
References:
https://www.ericsson.com/en/news/2026/8/ericsson-and-mediatek-raise-the-bar-for-location-accuracy
https://www.telecoms.com/5g-6g/ericsson-and-mediatek-put-precise-5g-positioning-through-its-paces
MediaTek overtakes Qualcomm in 5G smartphone chip market
MediaTek will use TSMC to make its Dimensity SoC’s in 2024
AT&T’s 600 MHz Deployment with Ericsson: Turning Low-Band Spectrum Into Coverage and Uplink Capacity
AT&T/Ericsson Demonstrate 5G-Based ISAC for Drone Detection at World Cup Stadium
Ericsson leads SK Telecom AI RAN vs. NVIDIA’s GPU centric AI RAN Alliance
AI-RAN and Agentic AI get real: Ericsson, Nokia, Verizon & other operators enter into a new network automation era
Analysis: Ericsson’s leading role in French INTENTION 6G project
5G in Europe: Broad coverage but 5G SA cloud native core network lags other major markets (Table)
Europe’s 5G market status appears increasingly uneven. The region has achieved broad 5G population coverage, but its transition to the 5G Standalone (SA) core network—and the cloud-native 5G Core required to realize 5G features and capabilities—remains materially behind several major markets.
A recent workshop report from the EU-backed European 5G Observatory highlights stakeholder concern that European investment remains weighted toward radio-access coverage rather than toward core-network modernization. The underlying Observatory assessment also characterizes EU deployment as still predominantly NSA, meaning much of the installed 5G radio layer continues to rely on an LTE/EPC anchor rather than a service-based 5GC architecture. Participants estimated that only 18% of European 5G investment is directed to the core network, compared with 40% in China and 34% in the United States, South Korea, and Japan. Although European 5G core investment reportedly rose 31% from 2024, the prevailing view was that the next phase of 5G deployment must prioritize SA-capable core infrastructure rather than further expansion of basic 5G coverage.
This distinction is technically significant. Much of Europe’s existing 5G footprint remains based on Non-Standalone (NSA) deployment, in which 5G NR radio access is anchored to the LTE/Evolved Packet Core domain. The 3GPP-defined SA architecture instead pairs 5G NR with the 5G Core (5GC), including its service-based architecture specified principally in 3GPP TS 23.501 and TS 23.502. A full SA implementation enables functions that NSA cannot support as comprehensively, including end-to-end network slicing, native 5G QoS flows, ultra-low-latency service support, exposure of network capabilities through APIs, and more flexible edge and enterprise-service integration.
Europe’s aggregate 5G investment level is reported at 50.6%, substantially above the proportion allocated to the core but below China’s 72.0%, the United States’ 62.0%, and Japan’s 58.0%. The imbalance helps explain why headline coverage statistics do not yet translate into comparably mature 5G SA availability or widespread advanced-service monetization.
The Observatory report indicates that basic 5G coverage now reaches nearly all EU citizens, corresponding to 96.8% of households overall and 88.9% of rural households. However, deployment of SA-capable sites remains limited: on average, only 21.6% of European base stations are reported to operate in 5G SA mode, compared with 36% in the United States and 35% in China.
The gap becomes more pronounced when considering commercial service availability. While 21.6% of European base stations may be SA-capable, commercial SA availability to end users is reported at only 2.8%. India provides a notable contrast: the report places its SA-capable base-station share at 10%, below Europe’s, but commercial SA availability at 50%. The comparison suggests that Europe’s challenge is not only deployment of SA-capable RAN and 5GC infrastructure, but also the operationalization, device enablement, service launch, and commercial scaling of SA offerings.
Stakeholders attributed the low European availability figures partly to the current concentration of SA deployment in enterprise and business-to-business applications. Germany reportedly has the EU’s highest proportion of SA-capable base stations, at 63.2%, yet commercial availability stands at only 2.5%. Austria has the highest reported EU SA availability, although that figure remains modest at 8.7%.
This enterprise emphasis is understandable. A 5G SA network provides the architectural basis for differentiated connectivity services specified across the 3GPP 5G system framework, including network slicing and dedicated QoS treatment. In particular, 3GPP TS 23.501 defines network-slice concepts and service requirements, while 3GPP TS 23.503 specifies policy and charging-control mechanisms that can support service differentiation. Private 5G networks, campus deployments, industrial automation, logistics, utilities, and critical-infrastructure applications are therefore among the most immediate candidates for SA-led value creation.
Workshop participants nevertheless agreed that broader 5G SA deployment, together with private-network expansion, will be necessary if Europe is to capture the full economic value of 5G. That requires investment not only in 5GC functions, but also in cloud infrastructure, transport capacity and synchronization, orchestration, security, operational automation, and interoperable exposure frameworks. In practical terms, the opportunity is to shift from a coverage-centric 5G model to one capable of delivering programmable, assured, and differentiated connectivity services.
The European Commission established the 5G Observatory in 2018 as an evidence-gathering and policy-support mechanism. The Commission reports that the EU had reached 75.3% harmonized spectrum assignment by 2025, with several Member States approaching or achieving full assignment. Spectrum progress is important, but it does not by itself ensure SA maturity: operators must still convert spectrum assets and broad NR coverage into commercially available 5GC-based services.
The investment challenge is substantial. A GSMA assessment published earlier this year estimates that Europe will require approximately $550 billion in mobile-network investment over the next decade, while operators may have access to only about $312 billion. That implies an investment gap of roughly $238 billion—one that could constrain Europe’s ability to close the SA, cloudification, and advanced-network-services gap with leading global markets.
Mr. Johannes Theiss, DG CNECT Head of Sector for Advanced Networking Technologies and Applications at the European Commission noted that 5G standalone (SA) will be crucial on the road towards 6G, though further work is needed to develop indicators that will be both meaningful and manageable. As preparations begin for how 6G progress will eventually be measured, the experience of tracking 5G provides a useful basis for identifying what worked well and what should be approached differently.
…………………………………………………………………………………………………………………………………………………………………………
5G SA -Europe vs RoW Comparison:
A 5G SA system requires both 5G NR radio access and a 3GPP 5G Core (5GC), rather than NR anchored to an LTE EPC as in NSA. The relevant architecture is defined principally in 3GPP TS 23.501 and associated procedures in TS 23.502. Accordingly, the table is best interpreted as an indication of how extensively each market has extended SA capability into the RAN—not as a direct count of deployed cores.
The EU’s 5G SA deployment footprint represented 20.9% of all mobile base stations in 2025, compared with 36.2% in the United States, 34.8% in China, 26.3% in Japan, and 26.2% in South Korea
5G SA deployment proxy – 5G SA base stations as a share of all mobile base stations (Source: Perplexity.ai):
…………………………………………………………………………………………………………………………………………………………………………
References:
Ookla: Uneven 5G deployment in Europe, 5G SA remains sluggish; Ofcom: 28% of UK connections on 5G with only 2% 5G SA
Ookla: Europe severely lagging in 5G SA deployments and performance
Dell’Oro: Telecom carriers are on a 5G SA spending spree with more to come
GSA: 5G Non Terrestrial Networks, 5G SA and 5G Advanced gain momentum
Dell’Oro: Mobile Core Networks +15% in 2025; Ookla: Global Reality Check on 5G SA and 5G Advanced in 2026
Dell’Oro: RAN market stable, Mobile Core Network market +14% Y/Y with 72 5G SA core networks deployed
AT&T deploys nationwide 5G SA while Verizon lags and T-Mobile leads
Impact of optical component shortages & bottlenecks explained + Hyperscaler’s CAPEX
LightCounting’s August 2026 market and component report highlights how severe physical supply shortages for high-speed optical components are reshaping data-center architectures and forcing cloud builders to expand their vendor pools from the traditional 2-3 suppliers up to 5-7 specialized hardware manufacturers. Qualifying and managing all of them is a new challenge.
- Applied Optoelectronics, Macom, and MaxLinear have already reported accelerating growth in Q2 2026 and we expect to see more examples in the upcoming earnings reports.
- Coherent and Lumentum have also reported improvements in growth rates for the last quarter, catching up with Eoptolink and Innolight (reporting at the end of August). Eoptolink has already disclosed a sharp increase in the profits expected for Q2. Accelink and CIG also report sharply higher profits.
- Tower Semiconductor reported triple-digit growth (y-o-y) in silicon photonics revenue in Q2-2026.
- GlobalFoundries plans to double its silicon photonics business in 2026.
- Cisco reported “a remarkable 28% y-o-y increase” in the networking segment revenues, up from 25% in Q1. Cisco reported three new hyperscale AI design wins and 40% growth in orders during the quarter. The company also reported record orders for campus networking – up 20% y-o-y.
- Arista Networks reported Q2 2026 revenue of $3.04 billion, marking its first-ever $3 billion quarter. This represented a roughly 38% year-over-year increase. Management also raised its full-year 2026 revenue growth outlook to 40%.
- Calix reported 21% y-o-y growth in Q2 and guided for another 15% in the current quarter. Growth in company’s revenue is attributed to sales of broadband equipment and AI-enhanced software to telecom operators.
- Extreme Networks reported 14% y-o-y growth in product sales, in part driven by success of its agentic AI networking platform.
Surging data-center traffic driven by AI queries has created acute optical component shortages, fundamentally altering supply chain qualifications for hyper-dense network switches and interconnects. The primary optical component bottleneck is a severe production shortage of Indium Phosphide (InP) laser chips and EML (Electro-Absorption Modulated Laser) components, which are failing to keep pace with a projected 53% surge in total optical transceiver and hardware demand, reaching $39 billion. Current demand for high-speed datacom optics is outstripping available supply by roughly 30%, forcing hyperscalers and system vendors to fundamentally restructure their supply chains. More details in the Addendum at the end of this article.
………………………………………………………………………………………………………………………………………………………………………………………………
Dell’Oro Group says that the rise of agentic AI and inference workloads is driving new demand and introducing network requirements that differ from those associated with training workloads. This shift is leading to significant expansion of front-end networks to support the broader buildout of AI infrastructure. We estimate that more than half of the growth in Front-end Data Center Switch sales over the next few years will be driven by AI-related opportunities. These deployments represent net-new expansion opportunities for both established vendors and new market entrants.
Accton, Arista, Celestica, Cisco, HPE/Juniper, H3C, Huawei, NVIDIA and others—emerge as the primary beneficiaries of this growth, according to the market research company.
“As AI infrastructure shifts from large-scale training to inference and agentic workflows, there is an increasing demand for general-purpose infrastructure, and expanded front-end network requirements,” said Sameh Boujelbene, Vice President at Dell’Oro Group. “The traditional assumption of a 10-to-1 ratio of XPU to CPU no longer applies across all deployments, with some environments moving closer to a 1-to-1 ratio. CPUs are becoming increasingly important for workloads orchestration and data movement. Additionally, networking for KV caching storage rack is also needed for inferencing applications,” added Boujelbene.
……………………………………………………………………………………………………………………………………………………………………………………….
As for the network providers/ISPs:
- Verizon announced a dark-fiber deal with Google to support the hyperscaler’s data center traffic growth. The deal is valued at over $1 billion and Verizon said there will be other deals announced before year-end.
- Deutsche Telekom said its AI factory for Germany’s public institutions and businesses that opened in February, has already sold the site’s 10,000 Blackwell GPUs capacity. DT is now looking at increasing capacity by another 20,000 GPUs.
- Comcast cites that its upstream broadband traffic is growing at 2.5x the rate of downstream data, which it links to AI-driven queries.
- AT&T is seeing rapid increases in large-scale data traffic requiring high-capacity metro and intercity fiber infrastructure.
Capex of Telecom Service Providers is still expected to be flat (or down 1%) in 2026, but AT&T and Comcast reported 16% and 20% y-o-y growth in capex for Q2, respectively. Both companies are investing more in broadband access. Verizon’s capex was also up, but only by 5% in Q2.
……………………………………………………………………………………………………………………………………………………………………………………….
Hyperscaler CAPEX:
The chart below illustrates the capital intensity of the TOP 4 Cloud companies. It is up sharply: Meta is already at 51% and Microsoft is at 46% in Q2. Oracle (not included in the figure) would have been off the chart at above 80% in Q1. The company reports earnings in early September and the key question is how it intends to finance future investments. Some moderation in spending growth is well needed. Amazon is the largest spender with $54.2 billion for the quarter, up 73% y-o-y. Amazon also increased guidance for 2026 capex from $200 billion to $220 billion.
Recent $ Trillion fundraising activities of Anthropic and OpenAI, supported by Nvidia, were widely covered yet both companies continue to lose money. Some financial experts refer to these activities as the “future for financial engineering.” Any innovation comes with some risk, but we all hope for the best.
………………………………………………………………………………………………………………………………………………………………………………………
Addendum: Optical Network Bottlenecks Explained:
-
- InP Epitaxial Production Limits: The primary physical constraint lies in the fabrication and epitaxy capacity for Indium Phosphide wafers used to build high-frequency lasers, where specialized manufacturing equipment (such as MOCVD systems) remains heavily backordered.
- Electrical Trace and Power Walls: Inside massive AI clusters, traditional pluggable optical modules face severe latency, thermal, and signal-integrity walls due to long copper electrical traces running from switch ASICs to the optical engine.
- Packaging and Testing Complexities: Shifting from legacy designs to dense 800G, 1.6T, and co-packaged optics (CPO) introduces intricate multi-die integration and hybrid bonding hurdles that slow down final module validation and packaging yields.
- Expanding Vendor Pools: Cloud builders are aggressively expanding their sourcing lists, moving from a concentrated pool of 2–3 legacy optical vendors out to 5–7 specialized component manufacturers to secure allocation.
- Ecosystem Pivot to Photonics: Major semiconductor players (such as Marvell’s acquisition of Celestial AI) and optical fabricators are pouring capital into silicon photonics and optical interconnect architectures to bypass standard copper and discrete laser constraints.
-
- Mechanism: LPO removes the power-hungry Digital Signal Processor (DSP) and clock/data recovery (CDR) chips directly from inside the optical transceiver module.
- The Fix: Raw, linear analog signals are driven straight from the switch or network interface card (NIC) ASIC.
- Benefits: Slashes module power consumption by 50% or more and cuts latency while preserving the hot-swappable, multi-vendor front-panel pluggable form factor.
- Trade-off: Requires host ASICs with advanced analog front-end and signal-equalization capabilities to process the raw electrical signals.
- Mechanism: CPO moves the optical engine (which converts electrical signals into light) off the front-panel cage and places it directly onto the same substrate or interposer as the switch or accelerator ASIC.
- The Fix: Electrical traces shrink from centimeters down to mere millimeters, completely bypassing lossy copper-clad circuit boards and high-power SerDes requirements.
- Benefits: Reduces optical-interface power consumption by up to 75% and maximizes bandwidth density for ultra-dense GPU scale-up fabrics.
- Trade-off: Reworks serviceability—if an optical engine fails, the repair domain shifts from a simple two-minute transceiver swap to board- or switch-level replacement
………………………………………………………………………………………………………………………………………………………………………
References:
Agentic AI and Inference to Supercharge Front-End Networks Growth, According to Dell’Oro Group
Networking chips and modules for AI data centers: Infiniband, Ultra Ethernet, Optical Connections
Goldman Sachs report: Optical Networking is the next mega trend in AI infrastructure
Cisco Execs: New “Network Supercycle” as Agentic AI Workloads Reshape Telecom Infrastructure
Cisco report: Agentic AI to reshape WAN traffic, AI inference will be ~25% of total traffic by 2035
Meta’s “Iris” AI Chip for MTIA: Implications for Telecom-Grade Optical Networking, DCI and High Capacity Ethernet Fabrics
Oriole Networks photonic networking platform to be integrated with AMD GPUs/CPUs for next-gen AI data center fabrics
Cheap Chinese AI Models: Unappreciated Threat to U.S. Hyperscaler AI Dominance
Huge Risks for the proposed $500B AI Investments from Giant Wall Street firms
Disclaimer: Perplexity.ai was used for research and analysis in this article.
……………………………………………………………………………………………………………………..
Executive Summary:
This past Monday, six giant Wall Street asset managers, private-equity firms and banks announced an effort to raise $500 billion to keep fueling the A.I. boom by financing more data centers, power plants and chips. The proposed platform would direct capital to Nvidia customers—including AI startups and data-center operators—at precisely the point when many have struggled to obtain funding through ordinary credit channels.
We take that as a huge warning sign for the proposed AI investments. Here’s why: If the underlying projects offered clearly proven cash flows, predictable utilization and collateral with durable value, lenders would not need a specially assembled consortium, headline-scale commitments and Nvidia’s direct involvement to make the loans happen. The initiative appears designed to overcome a financing bottleneck created by the extraordinary gap between AI infrastructure spending and demonstrated AI revenue.
This proposed $500 billion AI-financing initiative is less a validation of durable AI economics than an admission that the sector’s spending plans have outgrown its customers’ ability—or willingness—to finance them conventionally. Rather than demonstrating independently sustainable demand, the arrangement risks extending an investment cycle increasingly dependent on vendor-enabled credit, opaque commitments and financial engineering.
The firms—Apollo, BlackRock, Blackstone, Brookfield, Goldman Sachs and KKR—said they were working together to come up with that huge sum to lend to Nvidia’s customers, including the start-ups that use the company’s chips in data centers to develop and operate A.I. software. These customers, Nvidia said, have been struggling to secure financing for chips and data centers.
Nvidia will connect its customers with one of the six lenders, which will provide financing that could range from loans to credit. The financing will be “at attractive rates,” Nvidia said in a blog post.
In practical terms, this is a vendor-financing mechanism, even if the capital technically comes from third parties. Nvidia is helping its customers obtain the money required to buy Nvidia-dependent infrastructure. The more readily startups and data-center operators can borrow, the more equipment they can order; the more equipment they order, the more Nvidia can sell. That does not mean the demand is fictitious, but it does make it harder to distinguish independent end-user demand from demand supported by an ecosystem that is financing itself.
Circularity is the principal concern. When suppliers, investors, cloud providers, AI labs and lenders all have financial incentives to keep capital circulating within the same small group of counterparties, reported growth can look more robust than the ultimate economics justify. The risk is not simply that projects fail individually. It is that a shortfall in AI-service revenues, utilization or pricing could spread simultaneously through hardware vendors, AI developers, cloud operators, private-credit vehicles and the securities backed by their cash flows. Circular financing can blur the line between real external demand and investment-funded purchases.
Executives from the lenders joined Jensen Huang, Nvidia’s CEO & founder, for an unusual, extended interview on CNBC, where they talked up their new, seemingly insatiable desire to finance infrastructure for A.I. Huang said on the air. He added that “A.I. labs” and “A.I. start-ups” would have access to the financing. He did not name those companies or whether Nvidia would receive any money as part of the effort. David M. Solomon, Goldman’s chief executive, said the consortium was Mr. Huang’s idea.
“We need to raise this money as fast as possible,” said Larry Fink, BlackRock’s chief executive. He also stated: “There’s quite a bit of negativity around A.I. and data centers right now, but let’s be clear: This is going to be creating a huge amount of jobs.”
Skeptical Analysis:
The urgency in those statements deserves more scrutiny. Speed is not a substitute for underwriting, and job-creation claims do not establish that investments will meet their cost of capital. The industry is attempting to finance assets whose useful economic life may be much shorter and more uncertain than that of traditional infrastructure. A GPU fleet can lose competitiveness rapidly when a new architecture, memory standard or systems design emerges. Its resale value can fall sharply if capacity demand weakens. Treating such hardware as collateral comparable to a toll road, utility asset or long-lived building is a major assumption—not an established fact.
The announcement punctuates a head rush on Wall Street and in Silicon Valley into anything that even vaguely resembles A.I. The stocks of tech giants and chipmakers have soared for most of this year, and a pair of the biggest names in the space, Anthropic and OpenAI, are expected to file for initial public offerings that could value them at $1 trillion apiece.
But soaring equity valuations and enormous projected IPO valuations do not answer the basic return-on-investment question: who will pay enough, for long enough, to justify the total cost of the data centers, power plants, networking, GPUs and debt now being assembled? AI vendors may generate impressive revenue growth while still failing to earn enough to cover the depreciation, energy, financing and replacement costs of the infrastructure required to produce it.
Nvidia’s financing narrative appears to conflate broad interest in AI-enabled services with broad, independent demand for capital-intensive AI infrastructure. Governments, enterprises and startups may all seek access to AI capabilities, but the current demand for hyperscale GPU clusters, dedicated power capacity and purpose-built “AI factories” remains concentrated among a relatively small group of frontier-model developers and cloud platforms.
That distinction matters because broad adoption of AI applications does not automatically translate into economically sustainable demand for vast new data-center capacity. Enterprises can consume AI through APIs, hosted platforms, smaller models and open-source software without owning—or financing—dedicated high-density compute infrastructure. Nvidia itself describes demand as spanning enterprises, startups, governments, nations and AI labs, but the critical question for investors is how much of that demand converts into durable, contracted infrastructure revenue rather than experimentation, pilots or subsidized consumption.
As Bloomberg reported last week, OpenAI accounted for somewhere between 50% and 70% of Microsoft’s AI revenue in the 12 months to 30 June.
That degree of concentration is significant. Microsoft disclosed $24.1 billion in sales from OpenAI during the year ended in June; outside estimates place that at more than half, and perhaps about 70%, of Microsoft’s AI-related revenue. Much of the revenue reflects OpenAI’s spending on Microsoft cloud and model-development services, meaning a substantial portion of the apparent AI revenue base may be generated within a tightly linked commercial relationship rather than by a diversified population of external enterprise customers.
That could be problematic if the picture is similar at other hyperscalers and at other frontier AI companies like Anthropic, for example. A market in which a handful of loss-making model developers drive an outsized share of cloud and infrastructure revenue is inherently more fragile than one supported by a broad base of profitable end users. It exposes infrastructure providers and lenders to customer concentration, correlated capital spending and the possibility that a reduction in financing at one frontier lab quickly reduces demand across the supply chain.
Indeed, as has been reported by multiple correspondents – most notably staunch AI critic Ed Zitron – LLM makers like OpenAI are losing money hand over fist and can only charge so much for tokens before customers either curb their usage or switch to open source models. Zitron wrote:
“It’s estimated that 70% or more of the AI revenues of Microsoft, Google, and Amazon were from either OpenAI or Anthropic. UBS estimated that next year, Anthropic and OpenAI’s compute spend would be 48% of all Google Cloud revenues — which means that they likely account for even more than 70% of its AI revenues.”
The commercial challenge is not whether frontier models have value- they do. It is whether they can deliver that value at prices that exceed the combined costs of training, inference, electricity, networking, cloud capacity, depreciation and ongoing model development. If token pricing rises too far, customers may reduce usage, shift workloads to lower-cost models, or use open-source alternatives. If pricing remains low, frontier-model providers may struggle to cover their infrastructure bills. Open-source fine-tuning can materially reduce costs for specialized enterprise workloads, reinforcing the competitive pressure on premium proprietary-model pricing.
The most plausible downside is that the technology becomes broadly useful but insufficiently profitable to support today’s extreme capital intensity. That outcome would leave the sector with too much high-cost capacity, thin margins, declining GPU collateral values and lenders dependent on assumptions about utilization and cash flows that have not yet been tested through a downturn. The foundational financial risk is straightforward: AI-service revenues may prove insufficient to service the fixed obligations incurred to construct the infrastructure. Notably, Nvidia stock dipped modestly on Monday after The Financial Times reported that the company was nearing the mammoth financing deal.
While the contours of the arrangements were announced Monday afternoon, details remained scarce. A joint news release referred only to “memorandums of understanding” to “create dedicated pools of capital at significant scale.”
That language is important. Memorandums of Understanding (MoU’s) are not equivalent to binding, fully funded commitments. Until investors know the actual terms—capital committed, leverage permitted, collateral requirements, Nvidia’s role in losses, loan maturities, customer concentration limits and underwriting standards—the $500 billion figure is better understood as an ambitious financing target than as validated capital deployment. Reporting has described the initiative as a multiyear target rather than cash already committed, while questions remain about the scale of any Nvidia backstop.
During their television interview, the lenders’ executives alluded vaguely to “yield-based products” and even securitization, or the creation of bonds that would divvy up the revenue from A.I. labs into risky and less risky categories. There were several references to A.I. as a new asset class and to allowing smaller investors an opportunity to invest in debt backed by the data centers.
This is where the proposal moves from aggressive investment to potential systemic-risk creation. Securitization can distribute risk, but it does not eliminate it; it can instead diffuse difficult-to-value exposure across a wider investor base. Packaging AI-data-center debt into yield products may make financing more available, but it also risks obscuring the quality of the underlying cash flows, the degree of correlated exposure among borrowers and the vulnerability of rapidly depreciating hardware collateral. The more complex the capital stack becomes, the greater the danger that investors mistake engineered liquidity for genuine economic value.
Executive Quotes:
“NVIDIA has reached an important milestone. We began by building chips; today, we are helping create a new class of productive, investable infrastructure: AI factories,” said Jensen Huang, founder and CEO of NVIDIA. “In AI, compute is revenue. NVIDIA compute is uniquely suited for this role. It is broadly adopted, flexible across models and workloads, fungible and transferable across customers and operators, and continuously improved through CUDA software — extending its useful life and improving its economics over time. It is supported by a deep global ecosystem of developers, customers and offtakers. That is why we are bringing the world’s leading long-term capital providers together to independently underwrite AI infrastructure. These financing platforms will help customers access scarce compute at scale and build the DSX AI factories that will power every industry and country in the age of AI.”
“Modern compute has emerged as a scarce, mission-critical asset class with compelling investment characteristics that is positioned to drive significant long-term economic growth and productivity gains,” said Apollo President Jim Zelter. “The combination of NVIDIA’s proprietary technology ecosystem and Apollo’s flexible, long-term capital base provides a strong foundation to support the next stage of the AI buildout as part of the broader Global Industrial Renaissance.”
“The AI buildout will require unprecedented investment and a skilled workforce to turn that investment into the infrastructure that will help power future growth,” said Larry Fink, Chairman and CEO of BlackRock. “This partnership deepens our relationship with NVIDIA, including through the AI Infrastructure Partnership, and brings together NVIDIA’s leadership in accelerated computing with BlackRock’s ability to connect long-term capital to essential infrastructure. Together, we can help deliver the compute capacity that companies need to grow and create more jobs, supporting the continued growth of the U.S. and global economies, while creating attractive, long-term investment opportunities for our clients.”
“NVIDIA has created extraordinary demand for its compute through an intense focus on customer value and versatile technology,” said Jon Gray, President and COO of Blackstone. “We continue to be enormous investors globally across the NVIDIA ecosystem, and this announcement further underscores our confidence in their platform and the future of AI infrastructure.”
“As our strategic partner, NVIDIA is enabling us to scale AI factories. We are excited about further collaboration to build and fund the backbone of AI globally,” said Bruce Flatt, CEO of Brookfield. “With demand for large-scale AI compute growing significantly as adoption scales across industries, compute is fast becoming the essential layer of infrastructure and a core pillar of the Brookfield AI infrastructure strategy.”
“We’re in a pivotal moment of a historic AI investment cycle. NVIDIA’s full-stack platform is in high demand and uniquely positioned at the center of that global buildout,” said David Solomon, Chairman and CEO of Goldman Sachs. “Our investment and distribution roles reflect our confidence in NVIDIA’s leadership, and we’re excited for the new opportunity to create a market for credit backed by NVIDIA compute.”
“Compute has become a critical infrastructure asset. As we’ve scaled our approach to digital infrastructure, we’ve learned that delivery, not ambition, is the hard part. That’s why we’re excited to build on our strategic partnership with NVIDIA, a founding investor in Helix Digital Infrastructure, to bring together NVIDIA’s accelerated computing platform with KKR’s long-duration capital, infrastructure expertise and capital markets capabilities to turn growing demand into real capacity at extraordinary scale,” said Joe Bae and Scott Nuttall, Co-Chief Executive Officers of KKR.
“This is the very beginning — like what it was when I started in the mortgage-backed securities market in the 1970s,” Mr. Fink said. “I look upon this as a next future for financial engineering.”
That analogy should be treated as cautionary, not reassuring. Financial engineering can expand access to capital and spread risk efficiently when assets have transparent valuations, stable cash flows and conservative underwriting. It becomes dangerous when it is used to finance unproven revenue models, rapidly obsolescing assets and demand forecasts that must remain exceptionally optimistic simply to justify the initial investment.
Huge Risks Explained:
If AI demand stalls, Nvidia faces a sharp reversal in hardware demand and a potentially damaging credit overhang, while lenders could be left financing underutilized data centers secured by equipment whose value can decline far faster than conventional infrastructure. The common vulnerability is that the same uncertain AI revenue streams would be expected to support chip purchases, data-center leases, project debt and securitized investment products. Nvidia is especially exposed to risks:
-
Order cancellations and lower pricing power. Cloud providers, AI labs and startups would likely slow GPU orders, defer deployments or renegotiate capacity commitments. Nvidia could face weaker revenue growth, inventory risk and pressure on the high margins that have supported its valuation.
-
Vendor-financing and counterparty risk. If Nvidia is arranging or backstopping financing for customers buying its systems, a demand slowdown could turn what appeared to be hardware sales into indirect credit exposure. Customers unable to earn adequate returns from AI services may struggle to repay loans used to buy Nvidia equipment. The risk is magnified where the company’s commercial success depends on borrowers gaining access to financing in the first place.247wallst+1
-
Collateral impairment. GPUs are not durable, slow-depreciating infrastructure assets. A new chip generation, a shift toward more efficient models, or weak utilization can materially reduce the resale value of installed systems. If lenders rely on those systems as collateral, a default could leave them holding equipment worth substantially less than the loan balance—and Nvidia could face lower demand for both new and prior-generation products. Moody’s identifies rapid capacity expansion and fast-changing chips, cooling and computing architectures as sources of overbuilding and technology-obsolescence risk.moodys
-
Feedback-loop risk. A slowdown could create a negative cycle: AI customers reduce spending; Nvidia’s sales weaken; lenders become more cautious; financing availability tightens; customers cut orders further. Where vendors, customers and capital providers are financially intertwined, the decline can be more abrupt than a normal inventory correction.
-
Equity-valuation risk. Nvidia’s market value reflects unusually strong assumptions about the longevity of AI spending, margins and growth. A reassessment of those assumptions could compress the valuation sharply even if Nvidia remains profitable. BIS warns that disappointment in AI returns could trigger a sudden financing pullback and turn the capex boom into a prolonged investment bust.bis
Lenders’ exposure:
The most immediate lender risk is a mismatch between long-lived debt obligations and unstable, technology-dependent revenue. A data center might be financed over many years, but its GPU fleet may need continual upgrades to remain competitive—requiring additional capital expenditure before the original debt is repaid. Moody’s notes that this combination of increasing capital intensity, uncertain compute requirements and structured finance can pressure developers, landlords and investors through execution, renewal and refinancing risk.moodys
Construction and power risks:
A demand slowdown could arrive before projects enter service. That creates a particularly difficult situation for lenders because construction interest, cost overruns and power-reservation charges may accumulate before revenue begins.
Permitting delays, local opposition, water constraints and electricity-grid limitations further compound this exposure. Reuters reported that lenders increasingly treat project readiness—including approvals, permits and local community support—as a credit factor because delays can increase costs, jeopardize covenants and prevent projects from reaching revenue-generating operation.
System-wide scenario:
The more serious scenario is a correlated unwind:
-
AI applications fail to deliver enough monetizable demand.
-
AI labs and cloud providers reduce compute commitments.
-
Data-center utilization and expected rental income fall.
-
Borrowers cannot service or refinance project debt.
-
GPUs and related infrastructure lose collateral value.
-
Losses reach private-credit funds, banks, securitized vehicles, insurers and institutional investors.
-
New financing becomes unavailable, causing further cuts in infrastructure orders and Nvidia sales.
This would not necessarily resemble the 2008 banking crisis: some first-loss exposure sits outside regulated banks. But Chicago Booth research estimates that a severe re-rating of AI-related debt could still produce roughly $60 billion to $140 billion in realized credit losses, alongside large equity-market effects.
What matters most:
The decisive question is not whether AI is useful or whether data centers remain necessary. It is whether cash-paying end users will generate sufficient, durable revenue to justify the full cost of the hardware, power, real estate, construction and financing now being committed. If that answer is no, the sector may discover that it financed capacity—not returns.
……………………………………………………………………………………………………………………………………….
References:
https://www.nytimes.com/2026/08/10/business/ai-nvidia-lenders-500-billion.html
https://www.wheresyoured.at/dont-look-up/
Curmudgeon: Caveat Emptor: Huge Debt and Circular Financing Deals Dominate AI Build-Outs (07/23/26)
Merry-go-round of dog chasing its tail: Relationship between U.S. hyperscalers and private Gen AI companies
AI infrastructure spending boom: a path towards AGI or speculative bubble?
Expose: AI is more than a bubble; it’s a data center debt bomb
Will Google Cloud’s AI and data analytics revenue +TPU IP licensing income offset huge AI CAPEX to produce a decent ROI?
Amazon’s Jeff Bezos at Italian Tech Week: “AI is a kind of industrial bubble”
Big Tech AI spending binge results in massive job cuts!
AI spending boom accelerates: Big tech to invest an aggregate of $400 billion in 2025; much more in 2026!
FT: Scale of AI private company valuations dwarfs dot-com boom
Big tech spending on AI data centers and infrastructure vs the fiber optic buildout during the dot-com boom (& bust)
AI Data Center Boom Carries Huge Default and Demand Risks
Can the debt fueling the new wave of AI infrastructure buildouts ever be repaid?
Gartner: AI spending >$2 trillion in 2026 driven by hyperscalers data center investments
Will billions of dollars big tech is spending on Gen AI data centers produce a decent ROI?
Telefónica incorporates AI for businesses voice communications vs. 3GPP/ITU specifications
Executive Summary:
Telefónica España made an announcement this week which indicates that Voice could be an important AI monetization opportunity for telcos. The Spain based telecom group is positioning its business voice portfolio around a key differentiator: the ability to embed AI-enabled capabilities directly into conventional fixed and mobile telephony, without requiring enterprises to migrate users or workflows to a separate communications platform. It is incorporating generative AI features into its network for things like call transcription and summarization, which it says is will transform “every voice conversation into usable, structured and actionable information,” as week as virtual assistants on fixed-line and mobile.
Targeted at large enterprises, public-sector organizations, and mid-sized businesses, the enhanced portfolio is intended to shorten call-response times, increase the proportion of calls handled, and convert voice interactions into structured, actionable business information. Telefónica reports that the AI-enabled tools can reduce time spent managing calls by an average of 60%, enabling organizations to handle a higher volume of customer interactions.
Telefónica has integrated artificial intelligence across its business voice offerings—from basic mobile services to advanced PBX and cloud-based telephony platforms—as part of its evolution toward intelligent voice communications. The proposal incorporates generative-AI functions within the Telefónica network, including call transcription, automated summarization, and virtual-agent capabilities. These functions are designed to preserve information that might otherwise remain unstructured within voice conversations, while helping organizations reduce missed opportunities and improve operational responsiveness.
Javier Pascual, Director of Product, Pre-sales and Provisioning at Telefónica Spain, said:
“We are the only operator that offers intelligent transcription and summarization of calls over fixed and mobile voice, making us the best way for companies to access digital technologies. This pioneering solution, which integrates generative AI into standard telephony, allows our clients to summarize and transcribe calls, as well as integrate 100% of virtual agents using natural language, thus improving productivity and agility.”
Cross-Portfolio Intelligent Voice:
Telefónica’s approach spans enterprise, public-administration, corporate-mobile, and mid-market customer segments. It applies to traditional and cloud-based voice solutions, including:
-
Centrex IP, Telefónica’s converged fixed-mobile business voice platform.
-
Centrex 365, a Microsoft-based cloud voice offering integrated with collaboration tools.
-
Enterprise mobile voice services.
A core capability is AI-based transcription and summarization of calls. By transforming voice conversations into searchable and structured records, the feature can support knowledge capture, customer-service follow-up, compliance-related documentation, and analytics workflows.
The company is also introducing Centrex AI, a virtual-agent capability based on advanced language models. Centrex AI is designed to support next-generation generative-AI interactions across channels beyond voice and to integrate with customer business applications. The virtual agents are intended to interpret natural-language requests in context, automate repetitive interactions, and provide faster, more consistent responses.
Telefónica states that the platform supports more than 100 languages and can operate continuously, enabling 24/7 multilingual customer engagement.
Operational and Vertical Use Cases:
Telefónica reports that the AI-enabled capabilities can improve agent efficiency by as much as 60% by reducing time devoted to repetitive tasks. The company also cites potential increases of more than 10% in the number of interactions managed, reflecting improved call-handling capacity.
Initial use cases focus on healthcare, public administration, retail, and industrial enterprises:
-
In healthcare, a WhatsApp-based AI agent can schedule appointments, provide immediate confirmations, and support multilingual exchanges.
-
For municipal governments, voice agents can address common citizen queries in multiple languages and route calls to the appropriate department.
-
For automotive dealerships, virtual agents can help manage service appointments and customer inquiries related to vehicle sales.
By integrating generative AI functions into the existing voice network and service portfolio, Telefónica is seeking to extend intelligent automation to established telephony environments rather than treating AI communications as a standalone application layer.
…………………………………………………………………………………………………………………………………….
Editorial Analysis:
Telefónica’s offer as an operator-integrated, proprietary AI overlay on existing fixed/mobile and cloud voice services, rather than as a service defined by 3GPP or ITU. The business voice offering builds on standardized fixed/mobile voice and cloud-telephony foundations, while its generative-AI functions—call transcription, summarization, and virtual-agent integration—appear to be operator- and vendor-implemented capabilities. Current 3GPP work provides enabling mechanisms for AI/ML in 5G systems, whereas ITU-R’s AI-related IMT work addresses radio-network evolution rather than AI-enhanced enterprise telephony.3GPP.
Relevant 3GPP specifications:
3GPP’s AI/ML work is primarily focused on network and RAN optimization, AI/ML model transfer and lifecycle management, data collection, and interoperability. Notably, 3GPP has stated that it does not plan to standardize the AI/ML models themselves; it instead standardizes the supporting mechanisms and controls.3gpp
ITU-R versus ITU-T:
ITU-R: There are no ITU-R Recommendations specifically governing AI-based telephony, generative-AI call summarization, or virtual agents. This is consistent with ITU-R’s mission: spectrum, radio propagation, and IMT radio-interface frameworks. Its IMT-2030/6G work includes integrated AI and communication as a broad capability area, but that concerns wireless-system capabilities such as distributed training and inference—not enterprise voice-service features.
ITU-T: This is the more relevant ITU sector for AI telephony and conversational AI, although its work is still largely horizontal rather than specific to IMS/PSTN calling:
-
ITU-T F.748.46 (2025) specifies requirements and evaluation methods for AI agents based on large-scale pre-trained models. Its scope includes recognition, comprehension, dialogue, generation, and reasoning—capabilities directly relevant to virtual voice agents.
-
ITU-T E.AIQ, Framework for quality evaluation of conversational AI systems, is under study in Study Group 12. It proposes KPIs and an “AI Quotient” approach for assessing AI systems in relation to QoS and QoE.itu
-
ITU-T Y.3178 defines a functional framework for AI-based network-service provisioning in future networks.itu
-
ITU-T Y.3661 (2025) specifies an architecture and mechanisms for customer-oriented intelligent network operation, including AI-supported recognition of user intent; this is adjacent to, but not a telephony-service specification
References:
Telefónica incorpora la IA a todas sus comunicaciones de voz para empresas
https://www.telecoms.com/ai/telef-nica-upgrades-business-voice-services-with-integrated-ai
Vodafone Spain (Zegona), MasOrange and Telefonica in possible RANco joint venture
Telefónica and Nokia partner to boost use of 5G SA network APIs
Ericsson and O2 Telefónica demo Europe’s 1st Cloud RAN 5G mmWave FWA use case
Telefónica launches 5G SA in >700 towns and cities in Spain
Telefónica and Nokia partner to boost use of 5G SA network APIs
Enable-6G: Yet another 6G R&D effort spearheaded by Telefónica de España
Dell’Oro: Enterprise PON Deployments expected to increase 844% year-over-year
According to a new Dell’Oro Group report, “PON in the Data Center and Premise Advanced Research Report“ recently published, total 2026 Data Center PON equipment revenues are expected to increase 844% year-over-year (Y/Y), driven by hyperscalers looking to use the point-to-multipoint technologies to reduce the cabling and power consumption requirements of their out-of-band management networks.
“PON technologies are increasingly moving from traditional residential networks to enterprise and data center applications, providing additional growth opportunities for PON equipment providers,” said Jeff Heynen, Vice President of Broadband Access and Home Networking market research at Dell’Oro Group. “We see hyperscalers and enterprises, both large and small, increasingly deploying PON technologies for passive fiber distribution that is lower cost and that maintains its value far longer than traditional copper infrastructure,” added Heynen.
Additional highlights from the PON in the Data Center and Premise Advanced Research Report:
- Total cumulative spending on data center PON equipment from 2026 to 2030 is expected to exceed $3 billion, as hyperscalers, neocloud providers, and colocation providers all deploy PON for their out-of-band and infrastructure management networks.
- Enterprises are increasingly deploying Passive Optical LAN (POL) as the long-term benefits of increased speeds and lower operational costs outweigh the costs of deploying fiber in the building.
- Chinese operators continue to deploy tens of millions of master and subtended ONTs to deliver fiber-to-the-room (FTTR) services to their residential broadband customers.
………………………………………………………………………………………………………………………………………………………………………………
Editorial Analysis:
This extremely bullish forecast points to a potentially important new use case for PON: not as a replacement for the high-bandwidth, low-latency Ethernet fabric that interconnects servers and storage, but as an economical physical layer for the separate networks used to monitor, provision, and recover data-center infrastructure. In that role, a passive optical distribution architecture can consolidate fiber runs and avoid electrically powered aggregation equipment in parts of the management network, potentially simplifying expansion and reducing operational overhead.
Dell’Oro’s projected 844% year-over-year revenue increase should be read in the context of an early-stage market: the percentage reflects rapid adoption from a comparatively small base rather than an indication that PON will displace mainstream data-center switching. Nevertheless, the report’s forecast of more than $3 billion in cumulative 2026–2030 spending indicates that hyperscale, neocloud, and colocation operators are sufficiently interested to make data-center PON a material adjacent market for OLT, ONT, and ONU suppliers.
The enterprise opportunity is somewhat different. Passive Optical LAN can extend fiber deeper into commercial buildings, with optical terminals serving end-user areas rather than relying entirely on copper horizontal cabling and access switches. The principal trade-off is front-loaded installation complexity—especially where fiber pathways must be added or upgraded—against the prospect of longer infrastructure life, higher available access speeds, and lower energy use over the building lifecycle. Dell’Oro also includes enterprise/MDU POL and business FTTR applications in its five-year forecast coverage, suggesting that it views these segments as part of the same widening PON equipment ecosystem.
China’s large-scale FTTR deployments provide a useful volume counterweight to these specialized data-center and enterprise applications. Master and subtended ONT architectures enable operators to extend fiber connectivity from the residence gateway to individual rooms, creating another demand source for optical endpoints and related PON equipment. Together, these developments suggest that future PON market growth will depend increasingly on diversification beyond conventional residential FTTH—while also requiring vendors to address application-specific management, installation, and interoperability requirements.
The Dell’Oro Group PON in the Data Center and Premise Advanced Research Report includes 5-year market forecasts for PON Optical Line Terminals (OLTs), and PON Optical Network Terminals (ONTs) and Optical Network Units (ONUs) used in Data Center [Out-of-band management (OOBM), infrastructure management (DCIM)], Enterprise/MDU [Passive Optical LAN (POL), Fiber-to-the-room for business (FTTR-B)] , and Fiber-to-the room (FTTR) applications. To purchase this report, please contact us by email at [email protected].
References:
PON in Data Centers Expected to Grow at 52 Percent CAGR from 2026-2030, According to Dell’Oro Group
Dell’Oro: 2H2026 Data Center Capex to Accelerate due to massive AI Deployments
Analysis: Broadcom’s end-to-end 50G PON Edge AI portfolio with WiFi 8 support
Highlights of FiberConnect 2024: PON-related products dominate
Nokia and Google Fiber trial 50G PON – first in the U.S.
Nokia and Hong Kong Broadband Network Ltd deploy 25G PON
HKT is first to deploy 50G PON technology in Hong Kong
AT&T’s 600 MHz Deployment with Ericsson: Turning Low-Band Spectrum Into Coverage and Uplink Capacity
Executive Summary:
AT&T has selected Ericsson to supply 600 MHz dual-band radios for deployment of the low-band spectrum it recently acquired from EchoStar. The equipment choice is notable not simply because it activates new spectrum, but because it enables AT&T to introduce uplink-enhancing eight-receiver (8RX) technology across its low-band holdings for the first time. AT&T says it will disclose rollout timing and commercial-availability plans later.
—>By selecting Ericsson’s dual-band radios for its newly acquired 600 MHz holdings, AT&T is moving from spectrum ownership to the more consequential task of operationalizing low-band capacity across its entire RAN.
Spectrum Is Not a Network:
The announcement illustrates an elementary but sometimes overlooked fact of mobile-network economics: a spectrum license is not yet a network capability. It becomes one only when an operator has compatible radios, antennas, baseband and transport resources, site access, configuration, optimization, and a viable deployment program.
AT&T’s EchoStar acquisition delivered approximately 20 MHz of nationwide 600 MHz spectrum alongside 30 MHz of 3.45 GHz mid-band spectrum. The mid-band component could be put into service comparatively quickly, while the 600 MHz component required radio equipment that AT&T had not previously deployed at scale. That hardware gap makes the Ericsson selection an operational milestone rather than a routine vendor purchase.lightreading+1
Ericsson’s selected radios support both 600 MHz Band 71 and 700 MHz Band 14—the latter associated with FirstNet. This dual-band approach potentially lets AT&T refresh or augment low-band radio infrastructure while adding a new frequency layer, rather than treating the 600 MHz deployment as a stand-alone overlay.fierce-network
Why 600 MHz Matters:
Spectrum below 1 GHz is valuable principally for propagation rather than for peak throughput. A 600 MHz signal can generally cover a wider geographic area and penetrate buildings better than mid-band spectrum, which makes it especially useful for rural coverage, suburban edge coverage, indoor service, and reliability-oriented capacity layers. AT&T and Ericsson characterize the deployment as improving capacity, reliability, coverage, and overall network performance.ericsson+1
That does not mean 600 MHz should be viewed as a substitute for mid-band 5G capacity. With a nationwide block of roughly 20 MHz, AT&T’s 600 MHz spectrum is unlikely to match the raw throughput potential of wider mid-band channels. Its strategic value lies instead in extending a dependable coverage foundation to locations where higher-frequency spectrum either reaches fewer users per site or suffers greater indoor loss.
In this respect, the acquisition and radio deployment form a complementary two-layer strategy. The 3.45 GHz spectrum can add substantial metropolitan and suburban capacity, while 600 MHz strengthens the wide-area and in-building layer beneath it. The relevant measure of success will therefore not be a single peak-speed benchmark, but whether AT&T can improve user experience at the cell edge, indoors, and in markets where macro-site density is inherently limited.
The 8RX Uplink Dimension:
The technically distinctive part of the announcement is AT&T’s plan to deploy 8RX technology across low bands for the first time. In simplified terms, 8RX refers to use of eight receive paths at the base station. This can improve the network’s ability to receive signals transmitted by user equipment, improving uplink link budget, interference handling, and potentially uplink capacity or consistency. 8RX improves uplink because the base station has eight receive branches—rather than the four normally used for low-band FDD radios—to hear and separate signals sent by handsets. More receive branches improve receiver diversity, spatial interference rejection, and link budget, especially for weak uplink transmissions at the cell edge or indoors.
This matters because mobile usage is no longer overwhelmingly downlink-centric. Consumers upload high-resolution video, participate in real-time communications, share content, use cloud applications, and increasingly interact with AI-enabled services that may generate meaningful upstream traffic. Enterprises also depend on upstream performance for field video, surveillance, industrial devices, connected vehicles, and edge-to-cloud telemetry.
Low-band FDD deployments commonly use four receiver paths; moving to 8RX can therefore increase the base-station receive capability at the coverage layer where user devices are most likely to operate at weak-signal conditions. Light Reading reported that AT&T views the Ericsson radio selection as a means to introduce this uplink-enhancing configuration across its low-band spectrum, although performance results and deployment scope remain to be demonstrated in the field.
The important caveat is that 8RX does not create spectrum. Its contribution is to improve how effectively the network uses the uplink resources it has, particularly where coverage, interference, or link budget constrain the user’s transmitted signal. The realized benefit will depend on radio design, antenna implementation, spectrum configuration, device capability, traffic mix, and local RF conditions. See Addendum.
A Modernization Program, Not an Isolated Upgrade:
AT&T frames the effort as part of a wider network-modernization program. The company says upgraded Ericsson infrastructure has already delivered up to two-times faster average speeds in relevant areas, a 10% reduction in dropped or blocked calls, fewer slow-speed incidents, and lower uplink interference; these are operator-reported figures and should not be generalized to the forthcoming 600 MHz layer until independently validated.
The deployment also fits AT&T’s stated Open RAN direction. In 2023, AT&T said it intended for 70% of its wireless traffic to flow over open-capable platforms by late 2026, with Ericsson among the suppliers supporting its modernization path. The 600 MHz deployment itself should not be conflated with an Open RAN announcement, but it will take place within a RAN estate being progressively modernized for greater openness and flexibility.
What to Watch Next:
The immediate question is deployment execution. AT&T has not yet released commercial launch dates, market sequencing, site counts, or device-support details for its new 600 MHz spectrum. Those disclosures will determine whether this is primarily a targeted coverage investment, a broad nationwide low-band overlay, or a phased modernization program extending over several years.lightreading+1
The more substantive test will be whether AT&T can translate a high-value spectrum acquisition into measurable improvements in rural availability, indoor coverage, cell-edge performance, and uplink experience. The Ericsson selection is the necessary first step: it turns an underutilized spectrum asset into a deployable radio-network program. The ultimate value will come from execution at scale.
Addendum- “8RX” Explained:
In this AT&T/Ericsson deployment, 8RX means eight radio-frequency receive chains at the network side. The handset still transmits its normal uplink signal; the cell site has more antenna/receiver observations from which to recover it.
This differs from 8RX in a handset or fixed-wireless terminal, where it generally improves the device’s downlink reception. Here, the direction is reversed: the base station’s added receive capability benefits device-to-network traffic.
Uplink mechanisms:
-
Diversity gain: The eight branches experience somewhat different fading and multipath conditions. Combining them makes it less likely that a deep fade on one path causes decoding failure.
-
Array/combining gain: When receive paths can be coherently combined, the desired UE signal arrives with a stronger effective signal-to-noise-plus-interference ratio. In an idealized case, doubling the number of equivalent receive branches from four to eight can provide roughly 3 dB of additional combining gain, though the field result depends on antenna correlation, propagation, and implementation.
-
Interference suppression: More antenna observations give the receiver more spatial degrees of freedom to distinguish a desired UE from co-channel interferers. This can improve uplink SINR and allow more robust—or, when conditions permit, higher-order—uplink modulation and coding.
-
Better cell-edge operation: The uplink is often the limiting direction in wide-area low-band coverage because UE transmit power is tightly constrained. Improving the base-station receiver makes a low-power device more likely to maintain service from a building, rural location, or cell edge.
Why it is significant at 600 MHz:
600 MHz’s principal advantage is coverage: it propagates farther and penetrates buildings more effectively than higher-frequency spectrum. But broad downlink coverage can expose an uplink asymmetry—the device may receive the cell reliably yet lack sufficient transmit power for an equally strong return path.
8RX specifically addresses that asymmetry. AT&T says its Ericsson radios will enable uplink-enhancing 8RX across low bands for the first time, while Light Reading notes that low-band FDD networks such as 600 MHz have typically used 4RX radios and that 8RX had been relatively uncommon below 1 GHz.
Conceptually, with NN receive branches, the base station adheres to this equation:
References:
https://about.att.com/story/2026/att-ericsson-enhance-wireless-nationwide.html
https://www.lightreading.com/5g/at-t-puts-600mhz-to-work-with-ericsson-for-coverage-and-uplink
AT&T/Ericsson Demonstrate 5G-Based ISAC for Drone Detection at World Cup Stadium
AT&T and Ericsson boost Cloud RAN performance with AI-native software running on Intel Xeon 6 SoC
Analysis: AT&T 1Q-2026 results: increased fiber penetration, FWA momentum, D2D deals, and mobile/home internet bundles
Analysis: AT&T’s $250B network investment to advance U.S. connectivity
AT&T deploys nationwide 5G SA while Verizon lags and T-Mobile leads
Starlink Mobile: NTN–Terrestrial Convergence, Network Capacity, and the Limits of Disruption
Executive Summary:
Elon Musk’s strategic objective for Starlink extends beyond satellite broadband connectivity for remote, maritime, aviation, and enterprise users. SpaceX is positioning Starlink for a broader role in mobile communications through an integrated non-terrestrial network (NTN) and terrestrial-network architecture. That strategy could combine direct-to-device (D2D) satellite connectivity, satellite backhaul, selected terrestrial access infrastructure, and potentially a retail mobile service.
Financial markets have treated that prospect seriously. Shares of U.S. wireless network operators have been under pressure since SpaceX President Gwynne Shotwell described the opportunity for “Starlink Mobile” as “huge,” in a CNBC interview on June 12th. On SpaceX’s recent earnings call, Shotwell characterized the U.S. wireless market as generating $600 billion in annual revenue and stated, “and I anticipate us to be able to acquire quite a few of their customers because I think our service will be better.”
Editor’s Note: The combined annual revenue of AT&T, Verizon, and T-Mobile is closer to $360 billion, depending on the reporting period and whether total revenue or service revenue is used.
The timing, service architecture, spectrum strategy, device ecosystem, and go-to-market model for a SpaceX mobile offering remain unresolved. A central distinction is needed: direct-to-device NTN service, satellite backhaul, a terrestrial radio access network (RAN), and a consumer mobile service are related but technically and commercially distinct propositions.
The most plausible medium-term outcome is not wholesale replacement of U.S. mobile network operators (MNOs), but deeper NTN–terrestrial integration. LEO satellite systems can extend coverage, improve resilience, support mobility, and connect underserved areas. Dense terrestrial networks retain structural advantages for high-traffic urban service, indoor coverage, and busy-hour mobile capacity.
“The perceived risk of Starlink Mobile disrupting the U.S. wireless industry is greater than the actual risk in the next one to two years,” says Morgan Stanley analyst Sean Diffley.
Starlink’s satellite-broadband business serves 12 million subscribers and reportedly generates profit margins above 60%. Musk’s ambitions, however, extend beyond connectivity for aircraft, ships, and remote fixed locations. SpaceX plans a large third-generation Starlink constellation, supported by the anticipated launch economics of its fully reusable Starship system. Starlink V3 satellites are expected to materially increase satellite capacity relative to earlier generations.
However, aggregate constellation capacity should not be equated with mobile-network capacity in dense urban markets. The relevant question is how much spectrum, beam capacity, spatial reuse, and backhaul capacity can be delivered to a given area during the busy hour.
Image Source: Facebook
…………………………………………………………………………………………………………………………………………………
Starlink Mobile could develop along four, potentially overlapping, paths:
The proposed Starlink Mobile service should therefore be assessed as an integrated network proposition, not as a simple satellite replacement for a nationwide cellular network.
Why Terrestrial Networks Retain an Advantage:
The central technical constraint is spatial/spectrum reuse. Terrestrial mobile networks deliver high area capacity by repeatedly reusing licensed spectrum across dense grids of macro cells, small cells, and indoor systems. Capacity is further increased through sectorization, advanced antenna systems, massive MIMO, carrier aggregation, and high-capacity fiber or microwave backhaul.
A LEO satellite system has different strengths. It can provide wide-area reach, support mobility, offer resilience during terrestrial outages, and extend coverage to regions where conventional network deployment is uneconomic. Yet its radio resources are shared across comparatively large beam footprints. This creates more demanding link-budget, capacity, and spectral-efficiency constraints in high-density areas, particularly for indoor service and busy-hour traffic.
Terrestrial-cell coverage areas vary considerably with spectrum band, terrain, urban morphology, antenna configuration, and load. Nevertheless, the architectural distinction remains: terrestrial networks achieve high capacity by spatially reusing spectrum over relatively small geographic areas, whereas satellite systems must manage shared capacity over much broader footprints.
For this reason, Starlink V3 satellites may substantially improve overall system capability without eliminating the terrestrial RAN advantage in dense urban and suburban markets. The strongest near-term use cases for NTN are likely to include coverage extension, emergency communications, maritime and aviation connectivity, mobility, IoT, selected backhaul, and service in remote or underserved locations.
3GPP’s NTN work is specifically directed toward integration with terrestrial cellular and wireline networks, reinforcing the view that NTN will develop as part of a broader communications architecture rather than wholly separate from it. Analyss opinions:
- Daiwa Capital Markets analyst Jonathan Kees similarly argues that Starlink is “only a complement, not a substitute” for existing wireless infrastructure.
- BofA Securities analyst Michael Funk reaches a comparable conclusion: “Existing direct-to-device satellite technology [is] underdeveloped and largely complementary to terrestrial mobile networks,” Funk says.
Spectrum, Capital, and AI:
SpaceX has acquired valuable spectrum assets through its EchoStar transactions. The FCC approved the transfer of approximately 65 MHz of spectrum to SpaceX for next-generation direct-to-device use. This is strategically important, but spectrum ownership alone does not establish a nationwide facilities-based mobile network.
A fourth nationwide MNO would require much more than spectrum. It would require:
-
Radio-access infrastructure, including macro sites, small cells, antennas, radios, and power.
-
High-capacity backhaul and transport networks.
-
Mobile core-network functions, service assurance, cybersecurity, and operational-support systems.
-
Device testing and certification, distribution, retail channels, billing, customer care, and subscriber-acquisition spending.
-
Sustained capital investment to expand coverage and capacity as traffic grows.
SpaceX must also finance Starlink V3 production, launches, constellation replenishment, gateways, ground infrastructure, and continuing development of D2D and NTN capabilities. At the same time, the company is investing heavily in AI infrastructure and Starship. Reported second-quarter capex exceeded $18 billion, including approximately $15.83 billion associated with AI infrastructure.
This does not establish that SpaceX lacks the financial capability to enter mobile communications. It does establish a major capital-allocation challenge. AI infrastructure requires rapid, front-loaded investment; terrestrial mobile networks require multi-year and geographically distributed deployment; and satellite systems require recurring capital expenditure for replenishment and capacity expansion.
—->Please see detailed discussion of this issue in Addendum below Conclusions subhead.
The key strategic risk is therefore prioritization. If AI, Starship, and satellite-constellation investment continue to absorb the majority of incremental capital, Starlink Mobile may progress more slowly as an independent terrestrial build. It may instead favor targeted terrestrial deployment, satellite-enabled backhaul, wholesale arrangements, or an MVNO model.
Starlink Coverage Issues as per Iain Morris of Light Reading:
Starlink is estimated to have between 3 and 4 million subscribers in the U.S., mostly in rural areas where there is no fixed or mobile alternative. That is, after all, why someone would opt for satellite broadband. But a small cell network built on the rooftops of these 3 to 4 million customers – or several times that amount – is not going to provide the mobile coverage demanded by anyone who even occasionally travels a few miles from home. And inside the building, what advantage would a Starlink cellular service have over Starlink-supported Wi-Fi?
SpaceX cannot possibly hope to address this coverage problem without some kind of physics-defying technological breakthrough. According to figures shared by an authoritative source on the telco side, a small cell might get you coverage about half a kilometer from the basestation, while a rural macro site would take you up to 20 kilometers from it. In urban areas where higher-frequency spectrum comes into play, this radius would probably fall to about 2 kilometers, making small cells look more competitive. The trouble there is Starlink’s lack of subscribers in cities.
Other estimates do not improve the outlook. William Webb, an academic and analyst who previously worked for UK telecom regulator Ofcom, thinks Starlink might be able to achieve a coverage radius of 1 to 2 kilometers with roof-mounted antennas, “assuming a reasonably high gain cellular antenna is integrated into the Starlink terminal and oriented correctly,” as he wrote in a LinkedIn update. But mounting below roof level would reduce the figure to less than 1 kilometer, according to Webb.
“The biggest issue will likely be that overall coverage will be patchy, concentrated in areas where there are the most Starlink broadband subscribers,” he said. “These are likely to be rural areas where fixed broadband provision is poor. As a result, coverage will likely be overlapping in rural communities but sparse outside of these areas.”
Commercial Models and Possible Partnerships:
SpaceX may determine that a fully independent nationwide mobile network is not the most efficient near-term approach. An MVNO arrangement would allow SpaceX to purchase terrestrial capacity at wholesale rates while offering a consumer service incorporating Starlink capabilities. Such a model could provide near-national coverage more rapidly, though it would reduce SpaceX’s control over network economics, service differentiation, and quality-of-service management.
Several Starlink Mobile outcomes are plausible:
Why would AT&T, Verizon, or T-Mobile enter an MVNO or deeper partnership with a prospective competitor? The answer may be competitive strategy. The first MNO to secure a commercially favorable relationship with SpaceX could gain a relative advantage in coverage, satellite integration, customer acquisition, or enterprise services.
SpaceX is “going to have to partner with one of the mobile network operators like a Verizon or T-Mobile,” says RBC Capital Markets analyst Ken Herbert. “They’re going to have to find ways to work with other companies.”
T-Mobile already has a direct-to-cell/device (D2D) relationship with SpaceX. Verizon’s FWA business is more directly exposed to fixed-broadband competition from Starlink, while AT&T, Verizon, and T-Mobile each possess extensive spectrum, RAN, transport, and customer-service assets that would be difficult and costly for a new nationwide entrant to replicate.
Evaluating Starlink/Spacex Deployment Plans:
Starlink’s satellite-broadband business serves 12 million subscribers and reportedly generates profit margins above 60%. Musk’s ambition, however, is substantially broader than providing internet access to aircraft and oceangoing vessels. SpaceX plans to deploy as many as 100,000 third-generation Starlink satellites, supported by the anticipated economics of its fully reusable Starship launch system. Starship remains in the test phase, but SpaceX has stated that it is designed to place up to 150 metric tons in orbit per launch at approximately one-tenth the per-kilogram cost of Falcon 9. Starlink V3 satellites are expected to provide roughly 10 times the capacity of earlier-generation spacecraft.
SpaceX is advancing these plans at a time when the U.S. communications market is already experiencing cross-sector competitive convergence. Fifth-generation mobile technology enabled MNOs to expand fixed wireless access (FWA) offerings and compete more directly with cable operators for residential broadband customers. The resulting service bundles combine mobile subscriptions, home broadband, and, in some cases, video and value-added services. Cable operators, in turn, have expanded mobile offerings, generally using Wi-Fi offload, their extensive wireline footprints, and mobile virtual network operator (MVNO) arrangements with incumbent wireless carriers for wide-area cellular coverage.
This competitive equilibrium has modestly favored wireless operators, although neither sector has emerged unambiguously dominant. AT&T, for example, is expected to generate sales and earnings growth over the next several years, albeit below broader market growth rates. Comcast faces a more constrained outlook, with relatively flat expected revenue and earnings performance. The valuation gap reflects both differing growth expectations and the capital intensity of maintaining and upgrading nationwide communications infrastructure.
Wall Street has amplified concerns about SpaceX’s competitive impact. Early research coverage has broadly characterized the launch, satellite-connectivity, and artificial-intelligence company as a disruptive force. The average analyst price target for SpaceX stock is approximately $227, implying a valuation near $3 trillion—roughly five times the combined market capitalization of AT&T, Verizon, and T-Mobile.
The major U.S. MNOs also operate with substantial debt burdens. AT&T, Verizon, and T-Mobile collectively carry approximately $420 billion in net debt, against about $140 billion in expected 2026 EBITDA. “I would not for a second sleep easily if I had the investments that [wireless telecom providers] have, and the debt that they have, and Elon Musk working on a superweapon as it pertains to my business,” says Drew Cupps of Polen Capital.
Nevertheless, Starlink faces significant capital, engineering, and deployment challenges before it can offer a broadly substitutable mobile service. Daiwa Capital Markets analyst Jonathan Kees argues that the investment required to build a V3-based Starlink constellation capable of materially disrupting terrestrial operators is underestimated. He estimates that constellation deployment could require more than $100 billion, excluding investment associated with SpaceX’s AI initiatives. Starlink is “only a complement, not a substitute” for existing wireless infrastructure, says Kees.
The fundamental constraint is spatial reuse and capacity density. A terrestrial cell site can concentrate spectrum resources over a coverage area measured in a few square miles, while a LEO satellite beam may serve a much broader footprint. Satellite systems therefore face more difficult link-budget, capacity, and spectral-efficiency challenges in dense urban environments, particularly for indoor service and high-traffic applications. V3 satellites may substantially improve system capacity, but they do not eliminate the capacity advantage of dense terrestrial RAN deployments.
For that reason, BofA Securities analyst Michael Funk does not view SpaceX as an immediate, large-scale competitive threat to incumbent MNOs. While some projections assign LEO-based mobile services 15% to 35% of the U.S. telecommunications market, Funk regards those estimates as overly optimistic. “Existing direct-to-device satellite technology [is] underdeveloped and largely complementary to terrestrial mobile networks,” Funk says. That assessment is consistent with Starlink’s present international model, which emphasizes partnerships with mobile operators rather than wholesale replacement of their networks.
Conclusions:
Starlink Mobile is an important strategic development in the evolution of NTN–terrestrial convergence. It should not, however, be evaluated as a satellite-only replacement for nationwide cellular networks.
LEO systems can provide broad-area reach, coverage extension, resilience, mobility, selected backhaul, and direct-to-device services. Dense terrestrial RAN infrastructure retains decisive advantages in spatial reuse, indoor coverage, spectrum utilization, and busy-hour capacity. Starlink’s most credible competitive impact in the near and medium term is therefore likely to occur in selected coverage, mobility, resilience, backhaul, and fixed-broadband segments—not through immediate replacement of nationwide terrestrial mobile networks.
The central uncertainty is which architecture and commercial model SpaceX will adopt. Its concurrent commitments to AI infrastructure, Starship, satellite replenishment, and spectrum assets may favor a hybrid or partnership-led approach over a fully independent nationwide MNO build.
In that outcome, incumbent operators’ spectrum, terrestrial RAN, fiber backhaul, tower infrastructure, customer relationships, and regulatory authorizations remain strategically important. Rather than becoming obsolete, those assets may increase in value as NTN and terrestrial networks become more tightly integrated.
If satellite and terrestrial systems become increasingly interdependent, incumbent operators’ assets may become more valuable rather than less. Spectrum, fiber backhaul, tower infrastructure, distributed RAN assets, customer relationships, billing systems, and regulatory authorizations remain essential elements of an integrated non-terrestrial network (NTN) and terrestrial-network ecosystem. EchoStar’s share-price appreciation over the past year illustrates the strategic value that spectrum holdings can command, particularly as SpaceX and AT&T acquire additional spectrum assets.
The three national U.S. wireless operators collectively hold spectrum assets estimated to be worth more than $400 billion based on recent transaction benchmarks. That figure is approximately 80% of their combined market capitalization and is broadly comparable to their aggregate debt burden. Spectrum holdings alone do not establish an investment thesis, but they provide a meaningful asset-value foundation as satellite, terrestrial, and hybrid connectivity models continue to evolve. They aren’t going to sell their spectrum. They have businesses to run, and those businesses are reasonably valued and growing.
Addendum: AI Capex is a Significant Risk for Starlink Mobile:
AI-data-center investment is a significant capital-allocation risk to Starlink Mobile, but it is not, by itself, evidence that SpaceX lacks the financial capacity to enter mobile communications. The more credible concern is that AI, Starship, constellation replenishment, and terrestrial mobile deployment will compete for capital simultaneously—potentially slowing Starlink Mobile’s rollout, limiting network densification, or increasing reliance on an MVNO partnership.
SpaceX’s capital spending has risen sharply: reported second-quarter capex exceeded $18 billion, including approximately $15.83 billion directed to AI infrastructure. Management indicated that similarly elevated capex could continue over subsequent quarters. Meanwhile, Starlink profitability and connectivity operations remain an important source of funding for the broader AI strategy.
A facilities-based mobile network would add another exceptionally capital-intensive program:
-
SpaceX has already committed approximately $19.6 billion through its EchoStar spectrum transactions for up to 65 MHz of spectrum, including associated debt obligations. Thus, the issue is not whether it can acquire any spectrum; it has already done so. The issue is whether that spectrum portfolio is sufficient for a competitive nationwide terrestrial-mobile network, particularly in high-density markets.reuters
-
A fourth nationwide MNO requires far more than spectrum: site acquisition, radios, antennas, power, backhaul, core-network functions, device certification, retail/distribution, customer care, and sustained subscriber-acquisition spending.
-
Starlink must also fund V3 satellite production, launch capacity, gateway infrastructure, constellation replenishment, and the evolution of direct-to-device/non-terrestrial-network capabilities.
These programs have distinctly different investment horizons. AI infrastructure requires rapid, front-loaded spending to secure compute capacity; terrestrial mobile networks require multi-year, geographically distributed capex; and satellite systems require continuous capital expenditure to maintain and upgrade orbital capacity.
The likely consequence is not cancellation of Starlink Mobile, but a more capital-disciplined architecture than a conventional nationwide greenfield MNO build. SpaceX has said it intends to build terrestrial mobile services but has not disclosed projected mobile capex or a site-count plan; analysts have warned that a competitive direct-to-consumer national offering would be very difficult without an MVNO agreement providing broad baseline coverage. This produces three plausible outcomes:
Satellite connectivity can reduce backhaul requirements at selected sites and enhance rural, maritime, aviation, emergency, and resilience use cases. It cannot economically substitute for dense terrestrial spectral reuse in the highest-traffic urban areas. Therefore, AI-related capex pressure is particularly relevant because Starlink Mobile needs substantial investment precisely where the satellite component provides the least complete replacement for terrestrial infrastructure.
A key additional risk is strategic prioritization, not immediate insolvency. If SpaceX continues allocating the overwhelming share of incremental capital to AI data centers and Starship, Starlink Mobile may emerge as a hybrid service, wholesale platform, or MVNO-enabled offering rather than a fully independent fourth nationwide carrier. That outcome would reduce the near- to medium-term disruption risk to AT&T, T-Mobile, and Verizon. It would also reinforce the more technically credible view that Starlink is likely to be complementary to terrestrial mobile networks for years, even if it becomes a powerful competitor in selected coverage, backhaul, and mobility segments.
……………………………………………………………………………………………………………………………………………………………..
References:
https://www.barrons.com/articles/spacex-wont-destroy-telecomit-will-just-reinvent-it-79b88d1a
https://www.lightreading.com/satellite/spacex-small-cell-plan-serves-up-more-musk-madness
FT: SpaceX considering Starlink Direct-to-Consumer mobile service & terrestrial cellular network infrastructure in the U.S. (See Comments below the article)
Ookla: Starlink a viable competitor for hybrid 5G/NTN services due to network performance improvements and larger coverage area
Ookla: D2D satellite connectivity surged 24.5% during last 9 months; Starlink’s footprint expansion leads the way
US Mobile’s new bundle combines its multi-network mobile service with Starlink residential internet
Direct-to-Device (D2D) satellite network comparison: Starlink V2 (Starlink Mobile) vs “Satellite Connect Europe”
Blue Origin announces TeraWave – satellite internet rival for Starlink and Amazon Leo
Eutelsat hails EC’s IRIS-2 project to take on U.S. NTN providers
Executive Summary:
The European Commission (EC) has announced the next steps for the IRIS2 [1.] project after reaching an implementation agreement with the prime contractors, one of which is Eutelsat. The EC agreed terms with the SpaceRISE consortium – comprising Eutelsat, Hispasat and SES – for deploying the satellite system, which includes adding 66 more low-Earth orbit (LEO) satellites dedicated to defense, security and emergency services, and launching services from 2029.
Note 1. IRIS² Backgrounder:
IRIS² is the European Union’s secure-connectivity LEO satellite program: Infrastructure for Resilience, Interconnectivity and Security by Satellite. It is intended to provide sovereign, resilient communications for EU governmental users while also supporting commercial broadband and connectivity in underserved areas. IRIS² is planned as a multi-orbit constellation of 290 satellites, principally in low-Earth orbit (LEO) and medium-Earth orbit (MEO). Combining LEO’s lower latency with MEO’s wider coverage is meant to support secure, high-availability communications across Europe and beyond.
IRIS² was established under Regulation (EU) 2023/588, the Union Secure Connectivity Programme for 2023–2027. The Commission awarded a 12-year concession in December 2024 to the SpaceRISE consortium—SES, Eutelsat, and Hispasat—with major European aerospace, satellite, and telecom companies participating as subcontractors.
The current planning cited by ESA anticipates first satellite launches in 2029, initial services in 2030, and full operations targeted for 2031. In the interim, initial governmental capability is to draw on pooled national satellite resources through GOVSATCOM.
IRIS² is commonly described as the EU’s third flagship space program after Galileo and Copernicus. Its strategic rationale is European digital and communications sovereignty: reducing reliance on non-European satellite systems, improving resilience against cyber and physical disruptions, and sustaining a European space/telecom industrial base. It is not simply a European retail satellite-internet competitor to Starlink; secure governmental connectivity and strategic autonomy are central design objectives.
A good part of the motivation to build the IRIS2 satellite system is to reduce reliance on U.S. satellites internet providers (e.g. Starlink, Amazon LEO), especially for sensitive military and government communications.
Principal uses of IRIS2 include:
-
Secure communications for EU institutions, Member States, defence/security users, embassies, and critical infrastructure.
-
Crisis management, disaster/humanitarian response, border or surveillance-related applications.
-
Broadband, transport connectivity, satellite trunking, and service in terrestrial coverage gaps.
-
Integration with 5G/non-terrestrial-network technology; the Commission’s 2026 terminal-industrialization call explicitly includes 5G NR NTN-compliant modem/baseband capabilities.
……………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………
Eutelsat CEO Jean-François Fallacher said this is “a very important milestone” that moves the project on from design and preparation to an “operational phase” and confirms Eutelsat’s role as the program’s “LEO lead.” Fallacher shared a high-level view of some of the system’s new technology that is hoped to give it an advantage over larger U.S. rivals. He said the multi-orbit MEO-LEO constellation is “very advanced,” and will have “much more” capacity than satellite internet provider OneWeb has today.
He also confirmed it will be based on 5G non-terrestrial network (NTN) standards, which is “important.”
“We use 5G NTN technology. For IRIS2, we want to normalize the technology … This is also a way to fight against the American giants because today each of the technology used by ourselves on OneWeb (now owned by Eutelsat), by Amazon or by Starlink are full proprietary technologies,” he said. “Going to normalized technologies such as 5G NTN will also help [in] bringing the cost down [and] having antennas which will be cheaper and of a smaller size.”
Eutelsat, which also operates 31 geostationary orbit (GEO) satellites, gained its LEO constellation with the acquisition of OneWeb in 2023. It is a distant second LEO operator with around 650 satellites, compared to Starlink’s more than 10,000. They both operate LEO constellations, but Eutelsat has a “fundamentally different constellation with different target customers, business models and scale.” LEO connectivity accounted for 25% of Eutelsat’s total revenue of €1.23 billion (US$1.41 billion) for the fiscal year ending June 30, and LEO revenue grew 70% year-over-year.
Photo Credit: HJBC/Alamy Stock Photo
………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………..
For IRIS2, Eutelsat will design and operate 264 dual Mil-Ka/Ku band and 66 Mil-Ka band satellites. The company said it will invest €2.23 billion ($2.57 billion) in the shared infrastructure and €1.16 billion ($1.34 billion) in the commercial infrastructure (which includes Ku payload and commercial ground equipment), with spending phased from 2027 to 2034. According to its concession agreement with the EC, the company said it expects revenue from IRIS2 activity to be more than €10 billion ($11.6 billion) from 2032 to 2040. The Ka-only satellites are planned to start launching in 2029, while the commercial Ku LEO system is expected to be operational by mid-2032. Eutelsat also said it plans to add 229 satellites to OneWeb through 2034 in addition to 440 new first-generation satellites for an investment of €1 billion ($1.2 billion).
https://www.lightreading.com/satellite/eutelsat-hails-5g-ntn-to-take-on-american-giants
European Consortium 5G NTN transmission paves the way for standards based direct to device (D2D) connectivity
Non-Terrestrial Networks (NTN) Tutorial: Architecture, Spectrum, and Technical Foundations
Non-Terrestrial Networks (NTNs): market, specifications & standards in 3GPP and ITU-R
ITU-R recommendation IMT-2020-SAT.SPECS from ITU-R WP 4B to be based on 3GPP 5G NR-NTN and IoT-NTN (from Release 17 & 18)
From LPWAN to Hybrid Networks: Satellite and NTN as Enablers of Enterprise IoT – Part 2
Telecoms.com’s survey: 5G NTNs to highlight service reliability and network redundancy
InterSAT extends Pan-African satellite services via Ku-band on Eutelsat 70B satellite
Orange France satellite Internet based on Eutelsat Konnect VHTS satellite
France’s Eutelsat nears deal to buy UK satellite internet company OneWeb
Optus and Nokia’s pre-“6G” Trial in Australia: Upper 6 GHz May Be Widely Deployable
Executive Summary:
Australia’s Optus and Nokia have delivered one of the more credible pre 6G demonstrations yet: a live-field trial in Sydney that paired multi-gigabit speed with good coverage. The most notable result was the 3.5 Gbps peak download rate and the indication that upper 6 GHz could support a macrocell footprint comparable to today’s 5G 3.5 GHz network. The trial suggested the upper 6GHz band can cover roughly the same footprint as Optus’ existing 5G 3.5 GHz layer—an encouraging sign for lower-cost 6G upgrades.
TABLE 1. Optus–Nokia 6G Trial: Editorial Comparison of Technical Takeaways
(Adapted from reported trial results.)
Why this trial stands out:
In early 6G discussions, spectrum, coverage, and deployment economics are inseparable. Higher-frequency bands can offer more capacity, but they often demand denser networks and new site builds; that is precisely why the Sydney result matters.ericsson+1
Optus and Nokia’s trial suggests upper 6 GHz may offer a useful compromise: enough bandwidth for high throughput, yet enough propagation performance—when paired with advanced antenna techniques—to preserve broad-area coverage on existing infrastructure.telconews.com+1
What was tested:
According to the reported trial details, the teams used 200 MHz of upper 6 GHz spectrum, divided into two 100 MHz channels between 6,890 MHz and 7,090 MHz. Nokia’s proof-of-concept AirScale massive MIMO active antenna unit used 768 antenna elements and 128 transceiver chains at an existing Optus site operating alongside a live 5G network.telconews.com+1
That setup matters because it moves the conversation beyond lab conditions. A field trial on a live site is a better indicator of how upper 6 GHz may behave in real deployments, where interference, propagation, and network integration all shape performance.telconews.com+1
The bigger 6G implication:
The strongest signal from the trial is economic, not just technical. If operators can use upper 6 GHz with existing towers and familiar radio footprints, they may be able to introduce early 6G services without rebuilding their networks from scratch.telconews.com+1
That would be a major shift in how the industry thinks about 6G rollout. Instead of requiring an entirely new layer of dense infrastructure, upper 6 GHz could become a practical evolution path from 5G to 6G, especially for operators seeking capacity gains without a full civil-engineering reset.nokia+1
Spectrum policy:
This trial also lands in the middle of a broader spectrum-policy debate. The upper 6 GHz band is widely viewed as strategically important for future mobile networks, and results like this strengthen the case for allocating at least part of the band to licensed mobile use.ericsson+1
At the same time, the band remains attractive for other services, including unlicensed use cases. The Sydney trial does not settle that debate, but it does provide real-world evidence that upper 6 GHz is not merely theoretical: it can deliver both range and capacity under conditions that resemble operational deployment.
Quotes:
According to Sri Amirthalingam, Optus Chief Technology Officer, the achievement forms part of Optus’ ongoing network innovation program and its long-term work to help shape the technologies, standards and spectrum frameworks that will underpin the evolution from 5G to 6G.
“This trial is an important milestone in Optus’ long-term 6G research program and helps us better understand how future mobile networks can be designed to meet Australia’s unique connectivity needs. Achieving 3.5Gbps using upper 6GHz spectrum demonstrates the significant opportunity this band could provide for future networks as demand for data-intensive applications such as AI services, immersive video, cloud applications and connected devices continues to grow.
“As networks become increasingly AI-native, that enhanced capability will unlock a new generation of AI-powered experiences, from smarter digital assistants that can understand their environment, to more immersive education and real-time support for frontline workers. Our focus is on turning these innovations into meaningful benefits for Australians,” he added.
Mark Atkinson, Head of Radio Networks, Mobile Infrastructure at Nokia, says, “This trial with Optus demonstrates the potential of upper 6GHz spectrum to deliver the high capacity and performance that future networks will require, while also validating how existing infrastructure can evolve towards 6G.
“At Nokia, we see AI and connectivity becoming increasingly intertwined, with AI-native networks and AI-RAN enabling operators to deliver greater efficiency, better customer experiences and the platform required for AI-powered applications at scale. Through continued innovation and collaboration with Optus, we are advancing connectivity through AI-driven network innovation, helping shape the future of mobile communications and supporting Australia’s leadership in next-generation digital and AI technologies.”
Conclusions:
The Optus-Nokia result is not a commercial 6G launch, but it is a meaningful milestone. A smartphone-form-factor speed record is eye-catching; the more consequential finding is that upper 6 GHz may be deployable on today’s network footprint with far less infrastructure disruption than many expected. That would significantly improve the business case for upper 6 GHz. If network operators can reuse existing sites and achieve coverage similar to 5G 3.5 GHz, then the transition from trial to deployment could be less capital-intensive than many expected.
For network operators, regulators, and vendors, that is the kind of evidence that can shape both deployment strategy and spectrum decisions. If upper 6 GHz continues to perform this well in additional trials, it could become one of the most important bands in the transition from 5G-Advanced to 6G.
……………………………………………………………………………………………………………………………………………………………………………………………
Frequently Asked Questions:
What is massive MIMO and why does the antenna element count matter for upper 6GHz performance?
Massive MIMO (multiple-input multiple-output) is a technology that uses a large array of antennas at a base station to serve multiple users simultaneously in the same frequency resource, using spatial beamforming to direct signal energy precisely toward each device. Higher frequencies like upper 6GHz experience greater signal loss over distance than lower 5G frequencies — so more antenna elements are needed to compensate through more precise beamforming gain. The Nokia antenna in the Optus trial packs 768 elements into a proof-of-concept unit; commercial 5G radios typically use around 192. That difference in element count is the primary reason the Optus trial reached 3.5Gbps while Vodafone’s October 2025 trial with a less advanced antenna reached 2.5Gbps using the same 200MHz bandwidth.
Will upper 6GHz 6G services actually reach consumers without new towers being built?
The Optus trial’s outdoor coverage result suggests it may be possible — but only if the antenna hardware at existing sites is upgraded. The 768-element Nokia antenna compensates for upper 6GHz’s higher path loss through beamforming, matching the coverage footprint of a 3.5GHz 5G cell. Nokia’s CTO has previously confirmed that a 768-element array at 7GHz can fit in approximately the same physical enclosure as a standard 5G unit, because the higher frequency means each element is smaller. If that holds through production hardware, operators could upgrade existing sites rather than build new ones — a critical factor in the cost and timeline of any 6G rollout.
Why does Australia’s spectrum regulator have to decide about upper 6GHz, and what are the options?
The upper 6GHz band (6,425–7,125MHz) is currently under a spectrum embargo from ACMA, meaning no new licenses can be issued while it evaluates how to use the band. The core decision is whether to allocate upper 6GHz to licensed mobile networks (enabling 6G), to unlicensed Wi-Fi (enabling Wi-Fi 6E/7 at higher outdoor power), or to some sharing framework. Mobile operators argue they need the spectrum for future 6G. The Wi-Fi industry argues the same spectrum would dramatically expand outdoor Wi-Fi capacity. There is no technical path that gives both industries full access to the same frequencies simultaneously — ACMA will need to choose, and the Optus-Nokia trial results are now part of the evidence base it will weigh.
What happens globally if countries allocate upper 6GHz differently — mobile in some, Wi-Fi in others?
This is the central risk that the GSMA and mobile standards bodies have identified since WRC-23. If a significant portion of the world’s population — particularly the US, which has already dedicated the full 6GHz band to unlicensed Wi-Fi — does not align on upper 6GHz for mobile, device manufacturers will face a fragmented market: 6G handsets designed for global use cannot rely on upper 6GHz connectivity in all markets. The result would be regional rather than global 6G ecosystems, with separate equipment lines and higher costs. Australia’s decision, while one country among many, will contribute to the critical-mass calculation for whether the WRC-23 mobile identification becomes commercially viable or remains a regulatory aspiration without a unified device ecosystem behind it.
…………………………………………………………………………………………………………………………………………………………………………………………………………
References:
How NTIA “Call to Action for 6G Leadership and Security” might influence 6G/IMT-2030 standards and 3GPP specifications
3GPP approves timelines for Release 21 which will specify 6G RAN, Core and 5G Advanced
IMT-2030 (“6G”) Minimum Technology Performance Requirements for Radio Interface Technologies
ITU-R M.[IMT-2030.EVAL] & ITU-R M.[IMT-2030.SUBMISSION] reports: Evaluation & Submission Guidelines for 6G RIT/SRITs (6G)
Roles of 3GPP and ITU-R WP 5D in the IMT 2030/6G standards process
Dell’Oro: 6G RAN Capex to reach $500 billion by 2034 + Counterpoint
ABI Research: 6G Radio Installed Base by Region from 2029 to 2034
Analysis: Cohere’s $28M U.S. DoD FutureG ISAC contract; OTFS vs OFDM; 6G-NR/IMT 2030 RIT standards outlook
Analysis: Ericsson’s leading role in French INTENTION 6G project
Analysis: Nvidia’s rumored new 6G AI-RAN – likely features/functions and industry impact
Analysis: Nokia’s new AI-RAN platform and Standalone AI-RAN node with Nvidia GPUs
Ericsson and Intel collaborate to accelerate AI-Native 6G; other AI-Native 6G advancements at MWC 2026
NVIDIA and global telecom leaders to build 6G on open and secure AI-native platforms + Linux Foundation launches OCUDU
Nokia and Rohde & Schwarz collaborate on AI-powered 6G receiver years before IMT 2030 RIT submissions to ITU-R WP5D
AI wireless and fiber optic network technologies; IMT 2030 “native AI” concept
Comparing AI Native mode in 6G (IMT 2030) vs AI Overlay/Add-On status in 5G (IMT 2020)
SKT 6G ATHENA White Paper: a mid-to-long term network evolution strategy for the AI era
Verizon’s 6G Innovation Forum joins a crowded list of 6G efforts that may conflict with 3GPP and ITU-R IMT-2030 work
Highlights of 3GPP Stage 1 Workshop on IMT 2030 (6G) Use Cases
Ericsson and e& (UAE) sign MoU for 6G collaboration vs ITU-R IMT-2030 framework
