Dell'Oro Group: AI Infrastructure & Memory Costs Drive Q1 2026 Data Center Capex Surge

The first quarter of 2026 witnessed a substantial surge in data center capital expenditure, a trend largely attributed to the rapid expansion of AI infrastructure and escalating memory component prices, according to a recent report by Dell'Oro Group. This increase reflects the intense demand for computing resources necessary to power the next generation of artificial intelligence applications and services. Hyperscale cloud providers are at the forefront of this investment wave, not only expanding their general-purpose infrastructure to accommodate the growing demands of public cloud services and agentic AI workloads but also significantly boosting their specialized AI infrastructure deployments. The report indicates that these providers are also facing rising storage requirements directly linked to AI. Looking ahead, Dell'Oro Group forecasts an even more pronounced acceleration in capex growth during the second half of 2026. This anticipated increase is expected to be propelled by the widespread adoption of NVIDIA's next-generation Rubin systems, which are designed to handle massive AI workloads, and by ongoing refresh cycles for custom accelerator platforms utilized by hyperscalers. Beyond the major cloud providers, the report notes a growing adoption of AI infrastructure among select enterprise verticals and sovereign cloud providers. However, growth in these sectors is somewhat constrained by uncertainties regarding return on investment and the current readiness of their infrastructure. Despite healthy near-term demand, some spending might have been brought forward in anticipation of expected price increases later in the year. The global data center capex outlook for 2026 has been revised upwards to exceed $1 trillion, with significant contributions from the top four US cloud providers: Amazon, Google, and Meta. This robust investment underscores the foundational role of data centers in the evolving AI landscape.