According to KDC's Worldwide Quarterly AI Infrastructure Tracker, spending reached $89.7 billion in Q1 2026, up 33% year-over-year but flat sequentially as growth normalizes off a larger base. The defining shift was structural: ARM-based rack-scale GPU servers overtook x86 as the dominant accelerated computing platform. IDC raised its full-year 2026 forecast to $497 billion on hyperscaler capex growth and emerging non-GPU AI demand.
The Q1 2026 results confirm that AI infrastructure investment has moved well beyond initial proof-of-concept phases into a sustained, multi-year capital commitment cycle while the competitive battle has shifted from how much compute gets bought to which platform wins it. Enterprise technology buyers, cloud service providers, and national governments are making long term decisions about where to build, how much to spend, and which AI workloads to prioritize.
For vendors, this means a prolonged period of elevated demand across accelerated compute, high-performance storage, and supporting network infrastructure, but also a fast-moving contest for architecture leadership, as ARM rackscale platforms displace x86 in the accelerated server market. For enterprises, the data signals that AI capacity is becoming a structural cost of doing business at scale, that storage refresh deferred during the initial AI buildout can no longer be postponed, and that late movers risk falling behind on both performance and cost efficiency.
KEY TAKEAWAYS Regional performance was mixed. The United States remained dominant at $67.9 billion (75.7% of global spend, +30.3% YoY), though growth continues to moderate. China (PRC) returned to growth at $7.8 billion (8.7% share, +9.3% YoY). The Middle East & Africa remained the fastest-growing region (+233% YoY to $1.1 billion), followed by APeJC (+62% to $5.8 billion) and Western Europe (up to $5.1 billion). Accelerated compute remains the structural backbone, alongside a growing non-GPU AI-centric layer. Server spending represented 97.6% of total AI infrastructure value in Q1 2026, and within that, a growing share of AI-centric demand is landing on infrastructure that isn’t GPU-accelerated at all. AI orchestration tooling, data-pipeline workloads, and CPU-only inference clusters that hyperscalers are running as a cost-mitigation strategy alongside their GPU buildouts. Deferred storage investment is catching up. After redirecting budget toward GPU and AI server spend for the past one to two years and treating storage refresh as postponable, enterprises can no longer put those purchases off. Pent-up storage refresh is now landing on top of genuine AI-driven demand, reinforcing the urgency behind external storage strategy even as AI-centric storage remains a small share (2.4%) of total AI infrastructure value.
