Published: 09 November 2026
Author: Global Telecoms & Infrastructure Desk
Reading Time: 6 minutes
Executive Overview
The global telecommunications sector stands at a critical juncture. While artificial intelligence (AI) has dominated boardroom discussions and industry conferences for years, telecom operators are increasingly demanding proof of tangible financial returns. The era of experimenting with generative AI and abstract machine learning concepts is giving way to a pragmatic, business-driven approach.
In a comprehensive interview with Capacity, Shen Jianda, Vice President at ZTE, addressed this strategic pivot head-on. Shen outlined why telecommunications providers must transition away from speculative AI initiatives and instead concentrate on high-impact, practical deployments. From streamlining complex network optimization to driving crucial energy-efficiency measures, ZTE’s leadership is championing a framework where AI functions as a core economic and operational engine rather than a mere buzzword.
As operators grapple with surging data consumption, shrinking margins, and the impending architectural shift toward 5G-Advanced and 6G, Shen’s insights provide a timely roadmap. This article examines ZTE’s vision for AI-native infrastructure, the commercial realities of modern telecom deployments, and how operators can unlock sustainable revenue streams while drastically reducing their operational expenditure.
Detailed Chronology: The Evolution of Telecom AI Integration
To understand where the industry is heading in late 2026, it is vital to trace the rapid evolution of artificial intelligence within telecommunications infrastructure over the past half-decade.
Phase 1: The Era of Siloed Automation (2020–2022)
During the early rollout of 5G networks globally, AI was largely treated as an auxiliary tool. Operators deployed narrow, rule-based machine learning algorithms to handle specific, isolated tasks—such as predicting battery failures in cell towers or identifying minor anomalies in power supplies. These implementations, while valuable, operated in silos. They lacked cross-domain intelligence and failed to impact the broader business model, serving primarily as proof-of-concept projects rather than enterprise-wide strategies.
Phase 2: The Generative AI Gold Rush and the Reality Check (2023–2025)
The explosion of generative AI in 2023 sent shockwaves through the tech industry, spilling over heavily into telecommunications. Operators rushed to test large language models (LLMs) for customer service chatbots, marketing copy generation, and code writing. However, the heavy computational costs, latency issues, and hallucinations inherent in early models quickly exposed the limitations of applying consumer-grade AI directly to mission-critical telecom environments. Operators realized that generic AI tools could not solve the intricate physics of radio frequency (RF) propagation or the stringent reliability demands of core networks.
Phase 3: The Pragmatic Shift toward Network Intelligence (2026 and Beyond)
Entering late 2026, the industry has entered a maturity phase. As Shen Jianda emphasizes, the focus has shifted entirely to measurable business outcomes. AI is no longer viewed as a standalone consumer gadget integrated into a telecom portfolio, but rather as an intrinsic, foundational layer of the network architecture itself. Modern telecom AI is targeted, deterministic, and deeply embedded into radio access networks (RAN), core routing, and energy management systems.
Supporting Context & Metrics: The Imperatives of Modern Networks
Why is this pragmatic pivot toward AI essential right now? The answer lies in the converging pressures of soaring operational costs, environmental mandates, and unprecedented traffic demands.
The Energy Crisis in Telecommunications
Modern mobile networks carry exponentially more data than their 4G predecessors, yet energy consumption remains one of the single largest operational expenditure (OpEx) items for communication service providers (CSPs). Traditional networks consume vast amounts of electricity even during off-peak hours, maintaining baseline power levels across idle or underutilized cell sites.
- The Efficiency Gap: Manual power-saving configurations are too slow and rigid to keep up with fluctuating user traffic patterns.
- The AI Solution: By deploying intelligent energy-saving algorithms—such as those integrated into ZTE’s modern architecture—networks can dynamically shut down or scale back specific carrier components, symbol-level sleep modes, and deep-sleep states in real-time without impacting Quality of Service (QoS). Industry data indicates that AI-driven power optimization can reduce base station energy consumption by up to 30% to 40% during low-traffic periods.
Network Complexity and Autonomous Operations
With the ongoing commercialization of 5G-Advanced and the foundational research paving the way for 6G, network architecture has grown unfathomably complex. Multi-vendor environments, dynamic slicing, edge computing nodes, and massive MIMO antenna arrays mean that human network operations centers (NOCs) can no longer manually monitor and configure every network parameter.
Autonomous operations—often categorized on a scale toward Level 4 or Level 5 autonomous networks—rely entirely on closed-loop AI automation. These systems can autonomously detect degradations, predict congestion events, re-route traffic, and self-heal before subscribers notice any disruption.
Official Insights: Shen Jianda on the Future of ZTE’s AI Strategy
In his dialogue with Capacity, Shen Jianda provided a clear-eyed assessment of what telecom operators should prioritize. Rather than chasing fleeting technological trends, Shen argues that technology vendors and carriers must form a symbiotic partnership centered on tangible utility.
1. AIR RAN and Intelligent Radio Access
One of the cornerstone technologies highlighted by Shen is the integration of AI within the Radio Access Network (AIR RAN). Traditional RAN optimization requires tedious drive tests and manual parameter tuning by field engineers. AIR RAN leverages real-time data analytics and machine learning models at the edge to optimize beamforming, spectral efficiency, and interference coordination dynamically.
Shen noted that by embedding AI directly into the air interface, operators can maximize their existing spectrum assets. In an era where spectrum licenses cost billions of dollars, squeezing extra capacity out of existing bands via intelligent software upgrades delivers an immediate, undeniable return on investment (ROI).
2. Bridging 5G-Advanced and 6G Horizons
As the industry bridges the gap between mature 5G deployments and the distant horizon of 6G, AI serves as the continuous thread. 5G-Advanced introduces advanced capabilities such as passive IoT, extended reality (XR) support, and ultra-reliable low-latency communication (URLLC). Managing these diverse service requirements simultaneously requires an agile, intent-driven network.
Shen pointed out that the AI frameworks being deployed in 5G-Advanced networks today serve as the foundational training ground for 6G tomorrow. By mastering autonomous operations, predictive maintenance, and intelligent slicing now, operators are inherently building the native AI architecture required for the natively smart networks of the 2030s.
3. Monetization Beyond Connectivity
For over a decade, telecom operators have struggled with the "dumb pipe" dilemma—experiencing flat or declining average revenue per user (ARPU) while data consumption skyrockets. Shen emphasized that practical AI deployments must unlock new monetization avenues.
By utilizing intelligent network slicing combined with AI analytics, operators can offer differentiated, guaranteed-performance connectivity services to enterprise clients. Whether it is a smart manufacturing plant requiring ultra-low latency for autonomous guided vehicles (AGVs) or a remote healthcare facility needing secure, high-bandwidth feeds, AI enables operators to package connectivity as a dynamic, value-added service rather than a commodity.
Future Outlook: What Lies Ahead for Telecom AI
As we look past 2026 toward events like Datacloud USA and beyond, the trajectory of telecom AI is set to accelerate, provided the industry maintains its disciplined focus on practical outcomes.
The Shift Toward Intent-Based Networking
The ultimate goal of telecom AI is the realization of Intent-Based Networking (IBN). Instead of network engineers writing complex scripts or configuring individual routers and base stations, operators will simply state a business intent (e.g., "Ensure zero-latency video streaming for this stadium event while minimizing power consumption by 25%"). The AI-driven network will automatically translate that intent into real-time configuration changes across the entire end-to-end infrastructure, continuously monitoring and self-optimizing to fulfill the mandate.
Sustainable Infrastructure as a Competitive Advantage
Environmental, Social, and Governance (ESG) criteria are no longer optional for major corporations. Institutional investors, regulatory bodies, and consumers increasingly scrutinize the carbon footprints of telecom giants. By embracing the kind of pragmatic, energy-focused AI deployments advocated by ZTE and leaders like Shen Jianda, operators can achieve compliance with stringent global green mandates while simultaneously slashing operational expenditures.
Conclusion
The message from ZTE’s leadership is unmistakable: the honeymoon phase of telecom AI speculation is officially over. The future belongs to operators and vendors who can demonstrate clear operational efficiencies, unlock innovative enterprise revenue streams, and maintain rigorous cost discipline. Through intelligent infrastructure advancements like AIR RAN, automated network operations, and relentless energy optimization, the telecommunications industry is finally turning the promise of artificial intelligence into a concrete, profitable reality.
