Abhishek Singh, Technology and Customer Business Executive at Amdocs, with 20+ years in IT, telecom and network leadership.
getty
For two decades, the telecommunications industry has operated on a familiar premise: Build the pipe, push the data and scale the infrastructure. My previous writings explored how the convergence of fiber and 5G is laying the nervous system of a new digital era, and how agentic AI can accelerate network rollouts. But something more fundamental is happening now. AI is no longer just a tool that optimizes telecom networks; it is becoming the network itself.
We are witnessing the emergence of what I call the "intelligent network," an infrastructure that doesn't just carry data but understands it, anticipates demand and evolves without waiting for human instruction. This isn't a distant vision. It's what I see as the defining shift of our industry.
For years, operators invested in automation, scripts and rule-based workflows to reduce manual effort, which was necessary but limited. A script that restarts a failed node is automation. A system that detects conditions leading to failure, reroutes traffic preemptively, resolves the root cause and learns from the event is autonomy.
That distinction matters because operators increasingly view AI as a strategic capability rather than an efficiency tool. Nvidia's 2026 State of AI in Telecommunications survey found that 90% of operators report AI is increasing revenue and reducing costs, while 77% expect AI-native networks to launch before 6G even begins. The industry is crossing a threshold—from networks that follow instructions to networks that sense, reason and act.
From what I've observed across large-scale transformation programs at some of the world's largest carriers, the intelligent network rests on four pillars:
Traditional network operations centers (NOCs)—with their banks of monitors, rotating shifts and reactive troubleshooting—begin to reach their limits when managing millions of endpoints across 5G, fiber, edge and IoT environments simultaneously. The scale, speed and complexity of modern networks increasingly exceed what human operators can monitor and diagnose in real time.
AI-native operations address that challenge by embedding intelligence directly into the network fabric, correlating thousands of signals across RAN, transport, core and edge layers to detect anomalies and emerging issues before they escalate.
AI-powered digital twins now simulate thousands of rollout and upgrade scenarios before a single dollar of capital is deployed, evaluating impacts on customer experience, traffic evolution and competitive positioning. At Mobile World Congress 2026, operators demonstrated digital twins that don't just mirror the network's current state but predict how failures might propagate, resolving issues before they reach subscribers.
Since my last article, agentic AI systems have moved from pilots into live telecom environments, evolving from tools that recommend actions to systems that can execute multistep workflows with minimal human intervention. That transition is accelerating: a RADCOM survey found that 71% of operators plan to deploy agentic AI in 2026, targeting autonomous fault resolution, automated complaint resolution and predictive churn prevention. These agents don't replace human judgment; they amplify it at machine speed.
Instead of configuring systems manually, operators define desired outcomes—such as maintaining 99.99% uptime for an enterprise customer—and the network continuously determines how to achieve them as conditions change. This shift moves network management from execution to orchestration, creating opportunities for differentiation and enabling new revenue streams across enterprise services, healthcare, autonomous vehicles and industrial IoT.
The intelligent network sits at the intersection of every major technology trend: fiber-5G convergence, edge computing, private 5G and the early groundwork for 6G. Each amplifies the others. Dense fiber enables the low-latency backhaul that AI-native networks demand. Edge computing brings intelligence closer to users. Private 5G creates programmable networks optimized in real time by agentic AI. And 6G, with its focus on integrated sensing, communication and AI, will inherit the autonomous foundations we're building today.
The numbers confirm it: The Fiber Broadband Association projects a 2.3 times increase in total fiber miles by 2029. Global telco network cloud spending is expected to reach $24.8 billion by 2030. The agentic AI telecom market alone is projected to grow from $3.75 billion to nearly $12 billion by 2030.
The path forward demands honesty about challenges.
Trust And Transparency: AI systems making real-time decisions about critical infrastructure must be explainable and auditable. As Nokia's Sandeep Sasikumar noted, "What customers value is purpose-driven progress: AI-driven automation that’s explainable, trusted, practical."
Data Fragmentation: As many operators remain in low-automation maturity with fragmented, siloed systems, the intelligent network requires unified data architectures.
Workforce Evolution: Telecom professionals must evolve from network operators to AI orchestrators, understanding not just how networks work, but how to govern the intelligent systems that manage them.
Security At Scale: As networks become more software-defined and distributed, embedded AI monitoring and zero-trust architectures must be foundational, not bolted on.
The most important lesson I've carried across 20 years in this industry remains unchanged: Infrastructure drives innovation. But infrastructure is no longer just fiber in the ground and towers in the sky. Its intelligence is woven into every layer—sensing, learning, adapting and acting.
The operators who embrace the intelligent network are the ones that I believe will redefine what telecommunications means, transforming from providers of connectivity into orchestrators of intelligent, adaptive digital ecosystems.
Forbes Technology Council is an invitation-only community for world-class CIOs, CTOs and technology executives. Do I qualify?
此内容由惯性聚合(RSS阅读器)自动聚合整理，仅供阅读参考。 原文来自 — 版权归原作者所有。