AI Agent Competitive
Intelligence
Intelligence Journeys
AI Use Cases for Utilities
Start New Project →
Private Broadband for Utilities
Start New Project →

Automation

Network automation replaces manual configuration and operations with software-driven processes, ranging from scripted tasks to fully autonomous, self-optimizing networks. Driven by the complexity of 5G standalone, cloud-native architectures, and multi-vendor environments, operators are pursuing higher autonomy — often framed against the TM Forum’s autonomous networks levels — to cut costs, speed service delivery, and improve reliability. Automation increasingly relies on AI and closed-loop systems that sense, decide, and act with diminishing human intervention. For operators and enterprises, the practical path runs through incremental autonomy: automating well-defined tasks before attempting end-to-end self-operation. This channel covers network automation across RAN, core, and operations — including orchestration, closed-loop control, and the AI techniques behind autonomous networks — with analysis of where automation is delivering measurable results and how operators are progressing toward higher levels of autonomy.

Nokia and Deutsche Bahn launch 1900 MHz 5G FRMCS rail network

Nokia and Deutsche Bahn launch 1900 MHz 5G FRMCS rail network
Nokia and Deutsche Bahn have activated a commercial-grade 5G Standalone network on the 1900 MHz band to validate Future Railway Mobile Communication System (FRMCS) operations on live tracks. The partners have launched a 5G SA deployment using the 1900 MHz (n101) spectrum band on DB’s digital railway test field in the Ore Mountains (Erzgebirge), Germany. The network is built with Nokia AirScale radio equipment and an optimized, cloud-native 5G core, and it operates on moving trains on outdoor tracks. The setup includes built-in failover, self-healing, and real-time monitoring to sustain service continuity in mission-critical environments.

Siemens and TRUMPF unify IT/OT for AI-ready manufacturing

Siemens and TRUMPF unify IT/OT for AI-ready manufacturing
Siemens and TRUMPF are aligning digital platforms and machine-tool expertise to tackle the long-standing integration gap between enterprise IT and shop-floor OT—laying groundwork for AI-enabled, software-defined manufacturing. The partnership centers on open, interoperable interfaces that connect CNCs, robots, sensors, and enterprise systems without brittle, bespoke integrations. Digital twins of machines and lines—paired with standardized interfaces—let teams test control logic, validate process changes, and train AI models before they hit the floor. The companies are positioning their combined ecosystem as a credible path to “AI readiness” for motion-centric operations where latency, determinism, and safety are non-negotiable. An edge-first data fabric can normalize time-series, vision, and event data for low-latency decisions, while cloud services handle training and fleet-scale analytics.

Ericsson NetCloud Adds Agentic AI to Enterprise 5G

Ericsson NetCloud Adds Agentic AI to Enterprise 5G
Ericsson is embedding an agentic AI framework into its NetCloud platform to accelerate self-healing, intent-driven operations across private 5G, Wireless WAN, and SASE. Ericsson is evolving its AI assistant, ANA, from a prompt-based helper into a multi-agent system that can interpret high-level intents, plan workflows, and coordinate specialized agents to act across the enterprise networking stack. Ericsson’s rollout will be staged. A troubleshooting orchestrator is planned for Q4 2025 to handle high-frequency pain points such as offline devices and degraded radio conditions, with a projected reduction in downtime and support cases by more than 20 percent.

Telefónica Level 4 Network Automation Achievements

Telefónica Level 4 Network Automation Achievements
Telefónica is translating years of network automation into tangible Level 4 autonomous operations in targeted domains—an inflection point for service quality, cost, and speed at 5G scale. Under its Autonomous Network Journey (ANJ), Telefónica is aligning to the TM Forum Autonomous Networks framework and pushing selected processes to Level 4—closed-loop autonomy with minimal human oversight. The company reports a 70% reduction in flapping-related service impact and removal of manual work in these incidents, advancing this use case to Level 4 maturity. The operator cites 80% faster analyses for planning, operations, and optimization; a 40% drop in capacity issues; more than 90% reduction in sites experiencing high load with widespread customer impact; and a 5% latency improvement via virtual optimization prior to rollout.

Broadband Access Equipment Up 7% as Fiber and FWA Gain Share

Broadband Access Equipment Up 7% as Fiber and FWA Gain Share
The quarter’s growth underscores a resilient access capex cycle despite macro uncertainty, with fiber and fixed wireless access (FWA) deployments offsetting sluggish cable spend. Fiber PON platforms and 5G FWA customer premises equipment (CPE) drove the uptick, while DOCSIS infrastructure outlays fell 13% year over year on weaker Remote PHY Device (RPD) purchases and a slowdown in new virtual CMTS (vCMTS) licenses. The competitive center of gravity in broadband is shifting. Operators prioritizing XGS-PON rollouts and 5G FWA are growing faster and spending more, while cable operators are pacing upgrades and deferring some distributed access architecture (DAA) investments.

Small Cells Forecast 2030: 61M Shipments, Neutral Host & 5G SA

LA28 selects T‑Mobile 5G as Official Telecom Provider
The Small Cell Forum’s 2025 Market Forecast points to a market shifting from experimentation to scaled deployment, with enterprise demand and new business models driving a faster cadence. SCF forecasts cumulative small-cell shipments to reach 61 million units by 2030, supporting an installed base of roughly 54.4–54.5 million radio units and annual vendor/integrator revenues of about USD 4.23 billion. Indoor enterprise deployments continue to dominate, representing about 60% of rollouts in 2023–2024. SCF expects 5G SA small cells to grow at a 56% CAGR through 2030, with two-thirds of enterprise small cells co-located with edge compute by 2030.

LA28 selects T‑Mobile 5G as Official Telecom Provider

LA28 selects T‑Mobile 5G as Official Telecom Provider
T-Mobile for Business will serve as the Official Telecommunications Services Provider for the LA28 Olympic and Paralympic Games, positioning the event as a high-stakes proving ground for end-to-end 5G operations, broadcast connectivity, and fan experience at unprecedented scale. The LA28 organizing committee plans to run events across more than 110 connected locations, including over 40 competition venues distributed throughout Southern California. That footprint transforms LA28 into a distributed, city-scale network project where wide-area 5G must interoperate with venue networks, edge compute, and broadcast infrastructure under peak, dynamic loads.

Hitachi Rail Transforms Manufacturing with Private 5G from Ericsson & GlobalLogic

Hitachi Rail Transforms Manufacturing with Private 5G from Ericsson & GlobalLogic
Hitachi Rail’s Hagerstown factory is now powered by a secure Private 5G Network, thanks to GlobalLogic and Ericsson. This digital transformation enables smart manufacturing capabilities such as predictive maintenance, digital twins, AI-driven inspections, and real-time automation—positioning the plant as a benchmark for Industry 4.0 in North America.

Thailand NBTC frees 4.8 GHz for private 5G in factories

What Are Private 5G Networks? Benefits, Use Cases & How They Work
Thailand’s National Broadcasting and Telecommunications Commission (NBTC) plans to allocate 100 MHz in the 4.8 GHz range to factories and industrial estate operators to deploy non-public 5G networks under a private network operator (PNO) framework. The spectrum is to be granted on request and used solely for internal, non-commercial operations. Mobile operators may bid for PNO rights but cannot use this spectrum for public mobile service. The 4.8 GHz range sits within 3GPP Band n79, which means a relatively deep device and radio ecosystem that can lower total cost of ownership and accelerate time-to-deploy.

Lumen RapidRoutes 100G/400G Wavelengths with 20-Day Delivery SLA

Lumen RapidRoutes 100G/400G Wavelengths with 20-Day Delivery SLA
Lumen has introduced Wavelength RapidRoutes, a pre-engineered 100G/400G service with a 20-day delivery SLA aimed at removing months-long bottlenecks from enterprise and hyperscaler connectivity. The company is packaging pre-defined, high-demand optical paths as a catalog of ready-to-deploy waves, removing custom design cycles from many standard routes. Lumen’s RapidRoutes offers 100G and up to 400G wavelength services on prioritized intercity routes with an industry-forward 20-day service delivery SLA, shifting the customer experience from quote-engineer-build to select-provision-activate on pre-engineered paths. A portal-enabled experience with AI-driven tools and more than 300 automated workflows underpins ordering, change management, and capacity scaling.

Microsoft 365 AI integrates Anthropic Claude

Microsoft 365 AI integrates Anthropic Claude
Microsoft is preparing to license Anthropic’s Claude models for Microsoft 365, signaling a multi-model strategy that reduces exclusive reliance on OpenAI across Word, Excel, Outlook, and PowerPoint. According to multiple reports, Microsoft plans to integrate Anthropic’s Claude Sonnet 4 alongside OpenAI’s models to power Microsoft 365 Copilot features, including content generation and slide design in PowerPoint. This is a notable pivot from a single-model default to a best-of-breed approach that routes tasks to the model that performs best for a given function. For enterprises, especially in regulated and mission-critical domains like telecom, the shift implies more resilience, better accuracy for specialized tasks, and new options to optimize for quality, cost, and latency.

Cisco and NVIDIA Secure AI Factory for Agentic AI

Cisco and NVIDIA Secure AI Factory for Agentic AI
Cisco’s Secure AI Factory with NVIDIA, now integrated with VAST Data’s InsightEngine, targets the core blocker to agentic AI at scale: getting proprietary data to models quickly, securely, and at enterprise breadth. The new joint solution aims to collapse RAG pipeline delays from minutes to seconds, reduce integration risk with validated reference designs, and keep every interaction within security and compliance controls. By aligning Cisco’s AI PODs, NVIDIA’s AI Data Platform and DPUs, and VAST’s data intelligence layer, the offering provides a turnkey workload data fabric for production-grade AI agents. Cisco AI PODs now ship with VAST InsightEngine using NVIDIA’s AI Data Platform reference design, turning raw enterprise data into AI-ready indices and vectors in near real time.

Frequently Asked Questions

What’s the difference between automation and AI in a telecom context?
Automation, in its traditional form, generally follows pre-defined rules and scripted logic: a specific condition triggers a specific, predetermined response, with no real interpretation or judgment involved beyond what was explicitly programmed in advance. AI adds the capacity to learn from data, recognize patterns that weren’t explicitly anticipated, and make more nuanced decisions about what action makes sense in a given situation, even one the system hasn’t seen in exactly that form before. In practice, most modern telecom systems combine the two: AI analyzes a situation and decides what should happen, while automation infrastructure actually executes that decision reliably and consistently across the network, a combination often described as the foundation for agentic AI.
What is ‘zero-touch’ network operation, and how close is the industry to achieving it?
Zero-touch network operation describes the long-term industry goal of running networks with minimal direct human intervention, where the network itself handles configuration, fault recovery, and optimization automatically and continuously. It’s a meaningful aspiration rather than a fully achieved reality; the industry is progressing toward it in stages, with specific functions, like automated fault detection or dynamic capacity adjustment, achieving meaningful levels of automation well before the broader vision of an entirely self-managing network becomes reality. Standards bodies including ETSI have working groups specifically dedicated to defining the requirements for zero-touch network and service management, reflecting that this remains an active area of ongoing standardization rather than settled, widely deployed technology.
Why is automation considered essential for managing modern 5G networks specifically?
5G networks are substantially more complex than earlier generations for several compounding reasons: they rely heavily on virtualized network functions running across cloud infrastructure rather than fixed dedicated hardware, they support network slicing, meaning managing multiple distinct virtual networks with different performance guarantees simultaneously, and they often combine equipment and software from multiple different vendors rather than one integrated supplier. Traffic patterns also shift constantly and unpredictably as device density and application types continue to grow. Attempting to manually manage this level of complexity at the scale of a national or global network simply isn’t realistic, making automation effectively a practical requirement for operating a modern 5G network reliably at all.
How does automation relate to network orchestration?
Automation and orchestration are closely related but distinct concepts. Automation generally refers to executing a specific task without manual intervention, like automatically restarting a failed process or adjusting a configuration parameter. Orchestration refers to the broader coordination of multiple automated tasks and virtualized network components across their full lifecycle, deciding where different network functions should run, how they should scale, and how they interact with each other to deliver a complete service. In practice, orchestration systems often rely on underlying automation capabilities to actually carry out the individual tasks they coordinate, providing higher-level coordination while automation provides the lower-level mechanism for reliably executing decisions.
What are the risks of relying heavily on automated network systems?
Heavy reliance on automated systems introduces specific risks alongside its clear efficiency benefits. If an automated system makes an incorrect decision, that error can potentially propagate quickly and broadly across the network before a human notices and intervenes, compared to a manual process where mistakes tend to be more contained. There’s also a risk of reduced visibility and understanding among human staff over time, since heavily automated systems can create a gap between what the network is actually doing and what engineers fully understand about why, particularly as AI-driven decision-making becomes more involved. Operators generally manage these risks by maintaining careful guardrails for automated actions, expanding autonomous scope gradually as confidence grows.
How has automation changed the day-to-day role of network engineers?
Automation has shifted network engineers’ day-to-day work away from repetitive, manual configuration and troubleshooting tasks and toward higher-level responsibilities like designing automation policies, overseeing AI-driven systems, and handling more complex or novel problems that automated systems aren’t yet equipped to resolve independently. Rather than manually configuring each new service or personally diagnosing every fault, engineers increasingly spend time defining the rules, guardrails, and escalation criteria that govern how automated systems behave, then stepping in directly for situations that fall outside those parameters. This represents a meaningful skills shift, with growing demand for engineers comfortable working with automation platforms and AI systems alongside traditional networking expertise.
What’s ‘closed-loop automation,’ and why does it matter?
Closed-loop automation refers to systems that don’t just execute an automated action once, but continuously monitor the result of that action, compare it against the intended outcome, and adjust automatically if the result doesn’t match expectations, creating a self-correcting cycle without requiring a human to manually verify and re-trigger each step. A closed-loop system managing network capacity, for example, might automatically increase resources in response to rising traffic, then continuously monitor whether that adjustment actually resolved the congestion, and make further automatic adjustments if it didn’t. This concept is considered a meaningful step toward the zero-touch network vision, since it moves automation beyond simple one-off actions toward genuinely self-managing behavior based on real-world feedback.
AI Agent for Telecom Competitive Intelligence

Partner Hubs

Download content, access intelligence tools, and hear from executives.

Partner Events

  • Network X Vienna 2026
Scroll to Top