Will AI Impact Telecom's Bottom Line?

The Telecom Tower Infrastructure Shift 🌐🤖 In the AI era, telecom towers are rapidly evolving from passive infrastructure into active AI compute hosts. Traditional 5G architectures were never designed for modern AI workloads, creating a structural uplink deficit and shifting traffic patterns. To stay competitive, telecom operators and tower companies must adapt to these key macro trends: - Towers as Edge AI Nodes: Co-locating edge data centers at tower sites to meet strict AI latency demands. - AI-RAN Commercialization: Embedding AI directly into RAN hardware and software for real-time optimization. - Autonomous Network Operations: Shifting from manual management to real-time AI resource allocation (power, spectrum, and capacity). - The 6G Build Wave: New spectrum releases paired with AI as the killer app driving infrastructure densification. - Asia Leading the Pace: Operators like SoftBank, Indosat, Reliance Jio, and SK Telecom setting the global benchmark for smart towers. Industry Impact: This convergence is a structural transformation, not just an incremental upgrade. Organizations that successfully integrate edge AI at tower sites, deploy AI-RAN, and partner with hyperscalers will unlock entirely new revenue streams beyond traditional tenancy. The future belongs to distributed AI infrastructure nodes. How is your infrastructure strategy evolving to support the AI economy? #TelecomInfrastructure #EdgeAI #AIRAN #5Gto6G #Telecommunications #FutureTech #AIECONOMY

As AI infrastructure becomes increasingly recognised as critical national infrastructure, where will telecom operators create the greatest value? Join Robert Curran, Consulting Analyst at Appledore Research, and experts from Nokia, Nscale, Supermicro and Luth Computer Specialists Inc, as they examine how AI infrastructure is reshaping the telecom industry. ▶️ Watch the full discussion: https://lnkd.in/evqeWhPy Hear the panel's views on: • The role of telecom operators in the AI infrastructure ecosystem • AI factories, high-performance compute and edge inference • Sovereign AI and national infrastructure strategies • The commercial realities of power, utilisation and investment • Why partnerships are reshaping the market. 👉 Register to watch the full panel and receive a complimentary copy of TelecomTV's latest editorial report, Trends in Telco AI Infrastructure, featuring exclusive insights from leading operators SK Telecom, Swisscom, Telenor, Telia and Telus. #TelcoAI #AIinfrastructure #AIfactories #Telecoms #EdgeCompute #DigitalInfrastructure Elizabeth Hunt Nokia for service providers

The evolution of autonomous telecom networks is expanding beyond the RAN. Most of us are familiar with rApps for RAN optimization, but an equally exciting opportunity is emerging in the Core Network through cApps. These AI-driven applications have the potential to optimize cloud-native core functions, automate decision making, and improve service assurance across the 5G Core and future 6G Core. Here are five cApp concepts that I believe have strong potential: 1. Network Function Placement cApp Uses AI to determine the optimal placement of AMFs, SMFs, UPFs and other cloud-native network functions to reduce latency while minimizing cloud infrastructure costs. 2. Intent Validation cApp Validates operational intents against a digital twin before deployment, identifying conflicts and predicting their impact before changes reach the live network. 3. Core Energy Optimizer cApp Dynamically consolidates cloud-native network functions during periods of low demand, reducing energy consumption without compromising service quality. 4. Subscriber Experience Prediction cApp Correlates data from the RAN, transport and core to predict subscriber experience before issues are noticed by customers. 5. API Exposure Optimizer cApp Monitors Network API demand and automatically scales API infrastructure to maintain performance while controlling resource usage. The real opportunity lies beyond individual AI applications. Imagine rApps optimizing the radio network while cApps simultaneously optimize the core, working together to deliver true end-to-end autonomous network operations. As the industry moves towards 6G, domain-specific AI applications such as rApps and cApps could become fundamental building blocks for intent-driven, cloud-native networks. What other cApp use cases do you think will become essential over the next few years? #Telecom #5G #6G #AI #ArtificialIntelligence #AutonomousNetworks #CloudNative #CoreNetwork #5GCore #NetworkAutomation #NetworkAPIs #IntentBasedNetworking #DigitalTwin #Ericsson #Innovation

The next telecom revolution may not be about who builds the best network... It may be about who builds the network itself. For decades, the telecom ecosystem was relatively straightforward. Network vendors built the infrastructure. Operators deployed and operated it. Technology evolved generation after generation—from 2G to 5G—but the roles remained largely unchanged. AI is beginning to challenge that model. For the first time, the future network is no longer shaped only by telecom expertise. It is increasingly influenced by: AI platform companies Cloud providers Silicon innovators Software ecosystems Traditional telecom vendors This isn't just a technology shift. It's an ecosystem shift. The questions we should be asking are no longer limited to : • Who has the best radio? • Who has the most efficient scheduler? • Who delivers the highest spectral efficiency? They are becoming: • Who defines the AI platform on which future networks will run? • Where will intelligence be created and executed? • Who captures the value generated at the network edge? • Will the future network be telecom-first, AI-first, or a convergence of both? Perhaps the most important lesson from previous technology revolutions is this: The companies that define the platform often create more long-term value than those that build individual products on top of it. As our industry moves toward AI-Native RAN, Autonomous Networks, and eventually 6G, I believe the biggest disruption won't come from a new radio feature. It will come from the convergence of telecom, AI, cloud, and silicon into a completely new ecosystem. The future may not be built by one industry. It may be built between industries. What do you think? Will the next generation of networks still be defined primarily by telecom vendors, or will the center of gravity shift toward AI and cloud platforms? I'd love to hear perspectives from operators, technology vendors, hyperscalers, semiconductor companies, and fellow network architects. #Telecom #AI #AIRAN #6G #Cloud #EdgeComputing #NetworkArchitecture #Innovation #AutonomousNetworks #DigitalTransformation

Networks that think on their feet? 👀 Vodafone Albania and Nokia have shown how agentic AI could make 5G network slicing smarter, faster and more responsive. That means dedicated network capacity could be created exactly when and where it’s needed - whether that’s during an emergency, a packed event or a sudden surge in demand. Smarter networks. Stronger resilience. Better connectivity when it really counts. 🚀 Read more here: https://lnkd.in/e9PpHAqW #5G #AI #NetworkSlicing #AgenticAI #Connectivity #5GStandalone

🤖 Will AI finally give telecom networks a growth story? 📡 Ericsson's CEO says AI is expected to be an upside driver for mobile networks, lifting demand and the value of network infrastructure. The reasoning: AI applications, from agents to real-time inference on devices, need fast, reliable, low-latency connectivity, which could drive investment in and usage of advanced mobile networks. Telecom has struggled for years to monetise new capacity, so AI represents a potential new demand engine. Worth noting: this is a vendor's outlook, and the revenue case still has to prove itself. A real growth driver for telecom, or optimism from a network vendor? 👇 #AI #Telecom #Ericsson #TOAINews2026 #TimesofAI

Networks that think on their feet? 👀 Vodafone Albania and Nokia have shown how agentic AI could make 5G network slicing smarter, faster and more responsive. That means dedicated network capacity could be created exactly when and where it’s needed - whether that’s during an emergency, a packed event or a sudden surge in demand. Smarter networks. Stronger resilience. Better connectivity when it really counts. 🚀 Read more here: https://lnkd.in/dzY27ajC #5G #AI #NetworkSlicing #AgenticAI #Connectivity #5GStandalone

Networks that think on their feet? 👀 Vodafone Albania and Nokia have shown how agentic AI could make 5G network slicing smarter, faster and more responsive. That means dedicated network capacity could be created exactly when and where it’s needed - whether that’s during an emergency, a packed event or a sudden surge in demand. Smarter networks. Stronger resilience. Better connectivity when it really counts. 🚀 Read more here: https://lnkd.in/eQBHSvfd #5G #AI #NetworkSlicing #AgenticAI #Connectivity #5GStandalone

Nokia’s AI pivot is not really about AI. It is about escaping a 1% growth market. At its 2025 Capital Markets Day, Nokia showed investors three very different markets. Telecom providers: €72 billion addressable market by 2028. Expected growth: approximately 1% per year. Mission-critical and defence: €20 billion market. Expected growth: 11%. AI and cloud: Initially estimated at €24 billion. Expected growth: 16%. Five months later, Nokia increased its AI-and-cloud growth estimate from 16% to 27%. That is the entire strategy in four numbers: 1%. 11%. 16%. 27%. Nokia still earns most of its money from communications service providers. In 2025, operators generated approximately €15.3 billion—or 77%—of Nokia’s sales. But operator concentration has become a strategic trap. 5G investment has slowed. RAN is increasingly concentrated among a small number of vendors and customers. Operators demand lower prices, longer support periods and more integration accountability. And every major RAN contract carries enormous R&D, localisation, lifecycle and delivery costs. Meanwhile, AI infrastructure is creating demand for: • Data-centre interconnect • 800G and 1.6T connectivity • Ethernet switching • Coherent pluggables • Optical line systems • IP-over-DWDM • Automated network fabrics Nokia already owns many of those building blocks. The $2.3 billion acquisition of Infinera gave it greater optical scale, North American exposure, coherent technology and access to cloud customers. This is not Nokia trying to become an AI model company. It is Nokia betting that every AI factory will require an enormous network around—and increasingly inside—the data centre. The company is following bandwidth. And bandwidth is moving from connecting people to connecting compute. Part 3: If mobile is no longer the growth engine, what exactly happens to Nokia’s RAN business? #Nokia #5G #AIDataCenters #Optical #NetworkInfrastructure