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.

Google Cloud Study: 52% Deploy AI Agents, Driving ROI

Google Cloud Study: 52% Deploy AI Agents, Driving ROI
Google Cloud’s 2025 ROI of AI study signals a step-change: AI agents are now in production at scale and delivering measurable business outcomes. The study, fielded with National Research Group across 24 countries, finds 52% of executives report their organizations already use AI agents—specialized models that can plan, reason, and take actions. Momentum is material: 39% say their company has launched more than ten agents. Executives also report faster delivery cycles, with over half moving use cases from idea to production within three to six months, up from last year. Generative AI investment continues to climb as technology costs fall.

Arm and Meta Boost AI Efficiency with Neoverse

Arm and Meta Boost AI Efficiency with Neoverse
Arm and Meta have inked a multi-year partnership to scale AI efficiency from hyperscale data centers to on-device inference, aligning Arm’s performance-per-watt strengths with Meta’s AI software and infrastructure stack. Meta plans to run its ranking and recommendation workloads on Arm Neoverse-based data center platforms as part of an ongoing infrastructure expansion. The companies are co-optimizing AI software components—spanning compilers, libraries, and frameworks like PyTorch, FBGEMM, vLLM, and the ExecuTorch runtime—so models can execute more efficiently on Arm CPUs in the cloud and on Arm-based devices at the edge. The work includes leveraging Arm’s KleidiAI optimizations to improve inference throughput and energy efficiency, with code contributions flowing back to open source.

AWS US-EAST-1 Outage: What Went Wrong and How to Mitigate

AWS US-EAST-1 Outage: What Went Wrong and How to Mitigate
AWS experienced a major outage centered on its US-EAST-1 region in Northern Virginia, triggering cascading failures across dozens of cloud services and dependent applications worldwide. The incident began in the early hours of Monday and was initially mitigated within a few hours, though residual errors and recovery backlogs persisted through the morning in US-EAST-1. Engineering updates point to a DNS resolution problem affecting a key database endpoint (DynamoDB) alongside internal network and gateway errors in EC2, which then propagated across dependent services such as SQS and Amazon Connect. When a foundational component like DNS or an internal networking fabric falters, service discovery and API calls fail in bulk.

Capgemini completes $3.3B WNS acquisition

Capgemini completes $3.3B WNS acquisition
Enterprise demand is shifting from project-based consulting to managed, outcome-driven operations infused with AI and data. By combining WNS's scaled operations with Capgemini's consulting, engineering, and cloud capabilities, the company aims to capture this demand with end-to-end, AI-enabled "run and transform" offerings. The deal expands Capgemini's delivery footprint in India, strengthens its business services unit, and adds vertical platforms and playbooks that can be cross-sold to Capgemini's installed base in North America and Europe. For WNS clients, it opens access to broader transformation capabilities—cloud, data, and engineering—while preserving managed services continuity.

Jio 5G Nears 50% of Mobile Base

Jio 5G Nears 50% of Mobile Base
Jio closed the quarter ended 30 September with 234 million 5G users, up 86 million year-on-year and now approaching half of its 506.4 million total mobile base. Financial momentum tracked the subscriber and traffic surge. Jio Platforms posted quarterly revenue of INR 426.5 billion, up 14.9% year-on-year, and net profit of INR 73.8 billion, up 12.8%. Jio’s fixed wireless access service, Jio AirFiber, more than tripled year-on-year to 9.5 million subscribers. Bottom line: Jio’s 5G is now at meaningful scale with rising ARPU, heavier usage, and fast-growing FWA—setting up a monetization phase led by targeted pricing actions, application partnerships, and enterprise services as 5G-Advanced capabilities arrive.

T-Mobile Cyber Defense Center: 5G Security

T-Mobile Cyber Defense Center: 5G Security
T-Mobile has launched a purpose-built Cyber Defense Center alongside a new Executive Briefing Center, signaling a maturing, integrated approach to cyber resilience across its network and enterprise business. T-Mobile unveiled a centralized Cyber Defense Center at its Bellevue, Washington headquarters to detect, disrupt, and respond to threats in real time, complemented by an Executive Briefing Center that showcases industry use cases and a tie-in to the company’s always-on Business Operations Center for continuity during crises. T-Mobile’s Business Operations Center remains the operational backbone for network health, customer experience continuity, and coordinated disaster response, integrating data-driven dashboards that support rapid decisioning during natural disasters, outages, and high-impact events.

Boldyn and Apeiroon fast-track portable private 5G modules

Boldyn and Apeiroon fast-track portable private 5G modules
Defense, public safety, transport, and critical infrastructure need deterministic connectivity that moves with the mission. Traditional rollouts struggle with time-to-service, power, and backhaul constraints. Portable, “all-in-one” 5G modules help bridge that gap by putting the radio, core, and management closer to the edge, enabling local breakout, resilience, and consistent QoS. With 3GPP Release 16/17 features maturing and SA-first private networks becoming standard, demand is shifting from pilots to field-ready systems that can be mounted in vehicles, worn as backpacks, or staged in temporary zones.

Ericsson–EDC $3B deal accelerates 5G, Cloud RAN and AI

Ericsson Revamps Enterprise Wireless Solutions Partner Program
Ericsson has secured a three-year, $3 billion partnership with Export Development Canada (EDC) to expand R&D, fortify supply chains, and accelerate next‑gen network technologies with Canadian roots and global reach. The agreement arms Ericsson with EDC’s financing and insurance support to scale Canada-based projects in 5G, Cloud RAN, AI-driven network operations, and early quantum communications research while integrating Canadian suppliers into its international ecosystem. Over the term, Ericsson aims to deepen R&D executed across Ottawa, Montréal, and Toronto—where more than 3,100 employees work on 5G Advanced, 6G, quantum networking, and automation—expanding the country’s contribution to the vendor’s global product and standards roadmap.

Windows 11 Copilot: AI PC with Voice, Vision and Actions

Windows 11 Copilot: AI PC with Voice, Vision and Actions
Microsoft is weaving Copilot directly into Windows 11 so users can talk to their PCs and allow AI to see the screen and take actions, signaling a shift toward an “AI PC” model. Microsoft is rolling out a wake phrase so users can start tasks or ask for help hands-free, positioning voice alongside keyboard and mouse as a core input. Copilot Vision can view what’s on your screen - apps, documents, photos, even games—and provide step-by-step guidance or troubleshooting. Copilot Actions moves from advice to execution in a secure, contained desktop environment, while listing each step it takes. Windows 11 integrates Copilot directly into the taskbar, with one-click entry points for Voice and Vision.

AI-Driven Microwave Backhaul 2025: Fiber Split & 20 Gbps

AI-Driven Microwave Backhaul 2025: Fiber Split & 20 Gbps
Ericsson’s Microwave Outlook 2025 points to a backhaul market that will be almost evenly split between microwave and fiber by 2030, reshaping transport decisions for dense 5G and future 6G builds. Microwave already carries traffic for most live 5G networks worldwide, and a rising mix of E-band and emerging higher bands is closing the capacity gap with fiber for short- to medium-range links. For operators facing site densification, fiber lead times, and rising build costs, microwave provides a fast, resilient, and cost-optimized path to scale. E-band deployments are accelerating and overtaking legacy 38 GHz usage in several markets.

India–UK telecom innovation centre for 5G-Advanced and 6G

India–UK telecom innovation centre for 5G-Advanced and 6G
India and the United Kingdom have launched the India–UK Connectivity and Innovation Centre to accelerate secure, AI-driven, and resilient telecom technologies over the next four years. The two governments committed an initial £24 million—roughly ₹250–₹282 crore depending on exchange rates—to fund applied research, joint testbeds, field trials, and standards contributions in emerging telecom domains. The investment concentrates on three pillars: AI in telecommunications, non-terrestrial networks (NTNs) for satellite and airborne connectivity, and telecoms cybersecurity with open, interoperable systems. The multi-year window aligns to the critical runway for 5G‑Advanced and early 6G experimentation.

AI to Power India’s 6G Self-Healing Networks

AI to Power India’s 6G Self-Healing Networks
Telecom Secretary Neeraj Mittal underscored that AI will be central to the next generation of networks, not an add-on. The direction aligns with industry momentum: 5G-Advanced is already introducing AI-enabled RAN and core features via 3GPP, while 6G initiatives under the ITU-R IMT-2030 framework envision AI-native control loops, sensing-assisted connectivity, and tight integration of compute and communications. India expects 6G trials to begin around 2028, with commercial deployments to follow. Operators that harden their AI and automation capabilities during 5G-Advanced will enter 6G with a competitive execution advantage.

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