Table of Contents
An Omdia analyst told RCR Wireless News that AI adoption is now accelerating beyond customer-facing functions and becoming embedded in core network operations
In sum – what to know:
AI is moving deeper into networks – Telcos are expanding predictive and gen AI use cases from customer service to fault management, dynamic resource allocation, and real-time traffic forecasting, according to Omdia’s Inderpreet Kaur.
Public cloud adoption stays selective – Operators are shifting specific OSS and network applications to hyperscalers but remain cautious due to sovereignty, governance, and workload-portability issues.
Hybrid AI architectures are rising – Operators are mixing on-prem AI infrastructure with cloud-based training, while vendors like Red Hat and Nvidia compete to support modern containerized network platforms.
Telecom operators are preparing for a major shift in how networks are operated, automated, and optimized as artificial intelligence and gen AI take on a larger role across the industry, Inderpreet Kaur, senior analyst at Omdia, told RCR Wireless News.
She noted that AI adoption is now accelerating beyond customer-facing functions and becoming embedded in core network operations.
“Telecom operators are using AI (predictive and gen AI) capabilities across multiple use cases. While customer experience and contact centers have been at the forefront, AI use cases are now expanding in network management,” she said. Omdia’s latest research shows momentum in areas like fault management, predictive maintenance, dynamic resource allocation, and traffic forecasting. Models trained on performance indicators are already helping operators identify anomalies and adjust network resources in real time.
The rise of cloud-native functions is closely tied to this shift, but public cloud adoption in telecom remains gradual. While operators continue to explore AI and analytics in the cloud, several barriers persist, the analyst said. “Although challenges exist, operators are successfully using public cloud services to run various OSS and network-related applications,” Kaur noted. However, tightening rules on sovereignty and data governance mean telcos are taking more selective approaches.
Kaur emphasized that operators should take a workload-centric approach, placing control-plane and AI-heavy tasks in the cloud while keeping user-plane functions on-premises. Success, she said, depends on strengthening FinOps capabilities, ensuring workload portability, and embedding regulatory safeguards into vendor agreements.
As AI workloads become central to telco cloud strategies, Omdia expects a more diverse competitive landscape. Operators can pursue several architectural models: vertically integrated stacks from vendors like Huawei and ZTE; horizontal AI infrastructure built with partners such as Red Hat, VMware, HPE, or Nvidia; public cloud–based AI; or hybrid solutions that combine cloud training with on-premises inferencing.
Across all approaches, she said operators will increasingly require flexible deployment options, stable paths to automation, and cloud infrastructures designed to support both AI and traditional network functions. These trends will shape operators’ network-cloud roadmaps in the years ahead.
Global spending on telco network cloud infrastructure and software will rise from $17.4 billion in 2025 to $24.8 billion by 2030, according to a recent Omdia report.
The report noted that the expected surge reflects a 7.3% compound annual growth rate (CAGR) during the period, underscoring how operators are accelerating investment in cloud-native and AI-powered network transformation.
