Nokia: rapid surge in AI demand ‘testing’ existing US and European digital networks

A Nokia-commissioned study reveals that the rapid surge in AI demand is testing existing digital networks across the US and Europe, with infrastructure constraints threatening to limit AI growth and raise sovereignty concerns, prompting calls for urgent policy and investment actions.

Research commissioned by Nokia of roughly 2,000 business decision‑makers across the United States and Europe finds rising demand driven by an “AI supercycle” is already straining digital infrastructure and could sharply limit the scale of future AI deployments unless networks evolve quickly. According to the study, 88% of U.S. respondents and 78% of European respondents said infrastructure limitations could restrict AI’s scale, and many enterprises report downtime, latency and throughput constraints as live AI use increases. ^[1][2]

The research describes a fundamental reshaping of traffic patterns away from the traditional downlink‑heavy model dominated by browsing and video towards uplink‑intensive flows as edge devices generate vast volumes of data for upstream AI processing. Nokia’s own analysis of 5G behaviour warns that AI applications , from image generation to continuous voice assistants , can drive significant uplink spikes that existing 5G deployments were not optimised to carry. ^[1][4]

European business leaders in the survey were particularly blunt about readiness and sovereignty: 86% believe current networks are not ready for mass AI adoption and nearly 29% warned infrastructure constraints could force them to relocate AI workloads abroad, potentially undermining digital sovereignty. Security concerns are rising in parallel, with more than 80% of respondents across sectors saying AI introduces new risks and naming cybersecurity among the top AI use cases. ^[1][2]

Respondents and Nokia point to a set of policy and investment responses: regulatory simplification, faster access to spectrum, greater fibre rollout and targeted deployment of low‑latency edge infrastructure and private wireless for industrial sites. In the United States the study highlighted the need to optimise bi‑directional data flows, expand fibre capacity and accelerate edge rollouts if AI adoption is to continue apace. ^[1][2]

Industry data from Nokia’s related studies suggests a business case for such investments: enterprises deploying private wireless and on‑premise edge report rapid returns, with one study finding 87% of adopters achieved return on investment within a year and 94% of industrial users reporting reduced carbon emissions after combining edge and private wireless for AI workloads. The findings suggest upgrades can deliver both economic and environmental benefits while de‑risking mission‑critical AI deployments. ^[3][7]

The scale of the challenge is underscored by independent forecasting that projects explosive growth in AI agents and bandwidth demand over the coming decade, highlighting potential bottlenecks across access networks, edge gateways and interconnection points unless compute and network architectures co‑evolve. Speaking to Mobile World Live, Pallavi Mahajan, chief technology and AI officer at Nokia, said “The first wave of the AI supercycle has already reshaped industries and accelerated innovation”, and that “future waves will demand more advanced, AI‑native networks” where “connectivity, capacity, and low‑latency performance” are critical. Nokia is urging closer cross‑industry collaboration and more predictable regulation to enable timely network investment as AI demand accelerates. ^[5][1]

📌 Reference Map:

• ^[1] (Mobile World Live) – Paragraph 1, Paragraph 2, Paragraph 3, Paragraph 4, Paragraph 6
• ^[2] (Nokia newsroom release) – Paragraph 1, Paragraph 3, Paragraph 4
• ^[3] (Nokia newsroom release on ROI and edge) – Paragraph 5
• ^[4] (Nokia blog on 5G uplink traffic) – Paragraph 2
• ^[5] (arXiv forecasting paper) – Paragraph 6
• ^[7] (Nokia newsroom release on ROI and edge) – Paragraph 5

Source: Fuse Wire Services