Nokia (0HAF) has published its Global Network Traffic Report covering the period 2026 to 2035, forecasting that artificial intelligence will account for nearly half of all global network traffic within a decade. The report projects total network traffic will nearly quadruple to almost 3,789 exabytes per month by 2035, driven heavily by physical and agentic AI applications. The findings carry significant implications for network infrastructure investment, as Nokia identifies a fundamental shift in traffic patterns that will require redesigned, distributed networks capable of supporting both machines and people.

Key Points

  • Nokia Oyj (0HAF), a global leader in connectivity infrastructure headquartered in Espoo, Finland
  • Nokia's 2026–2035 Global Network Traffic Report forecasts AI will generate 44% of total global network traffic by 2035
  • Total monthly network traffic projected to reach nearly 3,789 exabytes by 2035; agentic AI traffic projected to reach 160 times its 2026 levels
  • Investors should watch how Nokia positions its distributed, AI-ready network infrastructure to capitalise on projected demand growth

Nokia Projects Total Network Traffic to Nearly Quadruple to 3,789 Exabytes Per Month by 2035

According to the report, global network traffic is expected to grow dramatically from 2026 levels, reaching almost 3,789 exabytes per month by 2035. Nokia attributes this expansion primarily to AI adoption across consumer, enterprise, and machine-to-machine applications. The report uses a bottom-up modelling methodology, combining application usage data, connected-device activity, and agentic workflows with third-party research, expert review, and Nokia's own laboratory testing. The scale of projected growth underlines the infrastructure investment demands that network operators and technology providers may face over the coming decade.

Physical and Agentic AI Applications Set to Reshape Uplink Demand and Low-Latency Requirements

The announcement highlights that AI traffic behaves fundamentally differently from traditional internet traffic. Physical AI — encompassing robots and connected vehicles — is projected to generate more than 90% of its traffic over the network uplink by 2035, reversing the historically download-heavy model that existing networks were built around. Agentic AI traffic is forecast to reach 160 times its 2026 levels, while physical AI traffic is expected to reach 105 times its 2026 level over the same period. Nokia's Chief Technology and AI Officer, Pallavi Mahajan, stated that networks built for traditional internet applications will increasingly have to support machines requiring real-time responses and substantial uplink capacity.

Edge and Metro Core Infrastructure to Serve 36% of AI-Native Traffic as Distributed AI Grid Emerges

Nokia's report forecasts a geographical shift in where AI data is processed. While the central cloud is expected to handle the majority of AI traffic, edge and metro core infrastructure is projected to serve 36% of all AI-native traffic by 2035. Nokia describes this distributed model as the "AI grid", driven by low-latency requirements, more efficient AI models, and data sovereignty considerations. Beyond AI, fixed wireless access is projected to generate more than one-third of consumer cellular traffic by 2035, further expanding the role cellular networks play in home broadband delivery alongside traditional mobile services.

Nokia's AI-Ready Network Strategy Positions the Company at the Centre of Connectivity Infrastructure Demand

Nokia describes itself as a global leader in connectivity infrastructure spanning fixed, mobile, IP, optical, core, and data centre networks. The publication of this report reinforces Nokia's strategic positioning within the AI infrastructure investment cycle, as demand for network upgrades grows alongside AI deployment. The company states it is actively building distributed, AI-ready networks to support the next phase of AI development. Investors may be watching how Nokia translates these projected traffic trends into commercial network contracts and infrastructure partnerships over the forecast period to 2035.