The global artificial intelligence (AI) infrastructure market is expected to exceed $100 billion by 2028, based on findings from the International Data Corp. (IDC). Spending on hardware for AI, including servers and storage, grew by 37% in the first half of 2024, reaching $31.8 billion.
“AI adoption is advancing as hyperscalers, cloud providers, private companies, and governments prioritize its integration,” Lidice Fernandez, group vice president at IDC, said in a media release. “At the same time, data centers are exploring energy-efficient designs to address rising concerns about power usage.”
Servers account for most AI infrastructure spending, making up 89% in the first half of 2024. Of this, 58% was for servers equipped with embedded accelerators, which enhance AI performance. IDC forecasts these advanced servers will dominate by 2028, representing over 60% of server-related AI spending.
The cloud continues to play a significant role, with 65% of server investments tied to shared environments like hyperscalers and digital service providers. In contrast, traditional businesses have been slower in adopting on-premises AI solutions.
Storage spending also grew, showing the massive data requirements for training AI models and managing datasets for inference processes. In the first half of 2024, storage investments increased by 36%, with over half tied to cloud infrastructure.
Regionally, the United States led AI infrastructure spending in the first half of 2024, followed by China, the Asia Pacific, and Europe. The Asia Pacific region is projected to grow the fastest, with a compound annual growth rate (CAGR) of 20%.
By 2028, servers in cloud environments are expected to represent 75% of AI infrastructure spending, highlighting the industry’s shift toward scalable and collaborative setups.