Research commissioned by Nokia found that rising demand driven by the AI supercycle is putting pressure on network infrastructure in the US and Europe, warning that existing connectivity will struggle to support the next phase of AI.<\/p>\n\n\n\n
The study, which surveyed around 2,000 businesses and decision-makers across the US and Europe, found networks must evolve rapidly to keep pace with increasingly complex AI workloads. In the US, 88 per cent of respondents said infrastructure limitations could restrict AI\u2019s scale, compared with 78 per cent in Europe.<\/p>\n\n\n\n
According to Nokia, AI is \u201credefining network requirements\u201d, with traffic patterns shifting away from traditional downlink-heavy consumer use towards uplink-intensive data flows. Applications such as autonomous vehicles, smart factories and remote healthcare generate large volumes of data at the edge that must be sent upstream for processing, placing pressure on networks originally designed for browsing and video streaming.<\/p>\n\n\n\n
In Europe, 86 per cent of enterprises believe today\u2019s networks are not yet ready for mass AI adoption. While two-thirds of enterprises surveyed already have AI in live use, more than half are already experiencing downtime, latency and throughput constraints. Security concerns are also rising as more than 80 per cent of businesses across sectors believe AI is introducing risks, with cybersecurity emerging as the top AI use case.<\/p>\n\n\n\n
Competitiveness is another pressure point. Nearly 29 per cent of European enterprise leaders warned infrastructure constraints could force them to move AI workloads abroad, potentially undermining the continent\u2019s digital sovereignty.<\/p>\n\n\n\n