TrendForce reports that North American cloud service providers are expected to keep investing in artificial intelligence infrastructure, and these continued investments are expected to increase global artificial intelligence server shipments by more than 28% YoY in 2026. The firm notes that the rapid growth of artificial intelligence inference services is boosting demand for general-purpose servers and is supporting both replacement and expansion efforts across data centers.
According to the research, TrendForce predicts that total global server shipments, including artificial intelligence servers, will accelerate from 2025, with a 12.8% YoY growth in 2026. This marks a shift from the recent period where the server market from 2024 to 2025 primarily centered on training advanced large language models, utilizing artificial intelligence servers equipped with GPUs and HBM to handle large scale parallel computing workloads.
TrendForce highlights a changing focus in the second half of 2025, when the growth of artificial intelligence agents, LLaMA based applications, and Copilot upgrades shifted cloud providers’ strategies toward inference services as a monetization approach. In this phase, artificial intelligence inference workloads are deployed not only on dedicated artificial intelligence server racks, but also on general-purpose servers that manage pre inference and post inference computing tasks alongside storage, broadening the infrastructure footprint required to support these services.
