FELICIA ITEJE
As the Middle East and Africa (MEA) accelerates its ambitions to become a global hub for artificial intelligence (AI), the evolution of telecommunications networks is emerging as a critical factor in supporting the region’s digital transformation. New projections from Nokia highlight the scale of this transition, forecasting that AI will account for 44% of global network traffic by 2035.
According to Nokia’s latest Global Network Traffic Report, total network traffic is expected to nearly quadruple from 2026 levels, reaching almost 3,789 exabytes (EB) per month by 2035. The findings point to a fundamental shift in network requirements as AI agents, autonomous machines, and connected devices generate increasingly complex traffic patterns.
The projections carry significant implications for MEA markets, where investments in AI infrastructure, hyperscale data centers, smart cities, and digital services are creating new demands on connectivity. As governments and enterprises expand their AI capabilities, telecom operators will need to support not only growing data volumes but also faster uplinks, lower latency, and real-time communication between machines.
Unlike traditional internet applications, which have largely prioritized downloading data, autonomous AI systems will depend on uploading information and exchanging data continuously. Nokia projects that physical AI applications, including robots and vehicles that interact with their surroundings, will generate more than 90% of their network traffic through uplinks by 2035.
Agentic AI traffic is also expected to increase 160-fold compared with 2026 levels, while physical AI traffic is projected to grow 105-fold over the same period. These developments will require networks capable of accommodating diverse workloads, from large-scale data transfers to continuous, low-latency connections.
Several Nokia partnerships across the region are already demonstrating how network infrastructure is evolving to accommodate more demanding digital applications.
In the UAE, Nokia has partnered with etisalat UAE to successfully launch the first commercial deployment of Nokia’s Cloud RAN solution powered by HPE carrier-grade servers in the Middle East. This landmark achievement showcases the continuous innovative drive of Nokia and etisalat UAE for future radio access networks, enabling the full potential of AI-RAN and demonstrating both companies’ commitment to delivering cutting-edge solutions that redefine network performance, efficiency, intelligence, and automation.
In Saudi Arabia, stc and Nokia achieved the MEA’s first commercial deployment of 5G Non-Standalone (NSA) Cloud RAN, which incorporates virtualized Distributed Units (vDUs) and Centralized Units (vCUs) across edge and far-edge data centers, bringing processing closer to users to help minimize latency. It also supports stc’s transition towards more programmable and automated networks, providing a foundation for future AI-driven RAN capabilities and the introduction of new services as network requirements evolve.
Nokia’s collaboration with Orange also highlights the growing focus on AI-native network infrastructure across the MEA. In April 2026, the companies announced a partnership with NVIDIA to develop and evaluate AI-RAN technologies, combining Nokia’s anyRAN 5G software with NVIDIA’s AI infrastructure. The initiative explores how AI can improve radio performance, spectral efficiency, and energy consumption through predictive optimization, while enabling new services such as integrated sensing and communication.
More recently, Zain KSA introduced Nokia Deepfield Cloud Intelligence across its network covering more than 100 cities in Saudi Arabia. The cloud-native analytics platform provides real-time visibility into network traffic and application performance, allowing the operator to identify potential issues, optimize network resources, and improve service reliability. The deployment also establishes an AI-ready data foundation to support the Kingdom’s expanding digital ecosystem.
Beyond radio access and transport networks, Nokia’s report highlights a shift in where AI workloads will be processed, with intelligence moving closer to users and connected devices.
Although central cloud infrastructure is expected to handle most AI traffic by 2035, edge and metro core infrastructure could serve 36% of all AI-native traffic. More efficient AI models, latency requirements, and data sovereignty considerations are expected to accelerate this decentralization, supporting the development of a distributed infrastructure known as the AI grid.
Beyond AI, Nokia forecasts that fixed wireless access (FWA) will account for more than one-third of consumer cellular traffic by 2035. The technology is expected to strengthen the role of cellular networks in delivering home broadband services alongside traditional mobile connectivity, offering additional opportunities for operators across the region.
Commenting on the findings, Pallavi Mahajan, Chief Technology and AI Officer at Nokia, said AI is changing not only the volume of network traffic but also the fundamental requirements of network infrastructure.
AI is changing what we need from networks, not just how much traffic they carry. Networks built for traditional internet applications will increasingly have to support machines that place new demands on the uplink and require real-time responses,” Mahajan said.
She added that the next phase of AI development will depend as much on network capabilities as on computing infrastructure, making network readiness a central consideration in AI planning. “At Nokia, we’re building the distributed, AI-ready networks that make this next phase possible,” she said.
