Gartner Press Release, “Gartner forecasts worldwide IT spending to grow 9.8% in 2026, exceeding $6 trillion for the first time,” Oct. 22, 2025. GARTNER is a trademark of Gartner, Inc. and its affiliates.
Ibid; Ross Kelly, “Global IT spending set to exceed $6 trillion in 2026,” ITPro, Oct. 23, 2025.
IDC, “Artificial intelligence infrastructure spending to reach $758Bn USD mark by 2029, according to IDC,” press release, Oct. 28, 2025.
Ibid.
World Semiconductor Trade Statistics, “Global semiconductor market approaches $1T in 2026,” press release, Dec. 2, 2025.
NielsenIQ, “Consumer Tech market growth estimate resets in 2026,” Jan. 2, 2026; Consumer Technology Association, “Global tech & durables market estimation – Q2 2025,” August 2025; NielsenIQ, “Consumer tech industry trends 2025,” April 1, 2025.
NielsenIQ, “Consumer Tech market growth estimate resets in 2026.”
Laura Allen, “CTA: Despite tariffs and economic headwinds, U.S. consumer tech revenue to hit $565 billion in 2026,” Consumer Technology Association, Jan. 4, 2026.
IDC, “Worldwide smartphone market to grow 1.5% in 2025, boosted by record Apple shipments in 2025 of 247.4 million units and 6.1% YoY growth, according to IDC,” press release, Dec. 2, 2025; IDC, “Global memory shortage crisis: Market analysis and the potential impact on the smartphone and PC markets in 2026,” Dec. 18, 2025; Dan Robinson, “Budget smartphones will be hit hardest as memory prices rise,” The Register, Jan. 15, 2026.
IDC, “Worldwide smartphone market to grow 1.5% in 2025.”
Jitesh Ubrani and Jean Phillippe Bouchard, “2025 holiday PC shipments exceed expectations as vendors accelerate inventory purchases amid supply concerns,” IDC, Jan. 12, 2026; IDC, “Global memory shortage crisis.”
Jitesh Ubrani, Ryan Reith, and Ramon T. Llamas, “Wearable devices market insights,” IDC, Oct. 21, 2025.
Linda Mann, “Wearables trends 2026: What to expect in the year ahead,” Cpon-Navi, Jan. 3, 2026; Ron Schmelzer, “CES 2026 makes one thing clear: AI’s next interface is you,” Forbes, Jan. 8, 2026.
IDC, “An investor’s guide to AI everywhere,” May 31, 2024; GIGALIGHT, “Global AI server market surges to $50 billion in 2023, expected to exceed 50% share by 2027,” Medium, Dec. 29, 2023.
IDC, “Worldwide server market revenue increased 61% during the third quarter of 2025, according to IDC,” press release, Dec. 11, 2025; Lyle Smith, “Global server market sets new revenue record in Q3 2025 as AI infrastructure spending accelerates,” StorageReview, Dec. 15, 2025; IDC, “Servers market insights,” Oct. 21, 2025.
IDC, “Servers market insights.”
Ibid.
Deloitte analysis of brokerage reports, analyst reports, and publicly available industry reports.
Deloitte analysis of industry research, company earnings materials and transcripts, and thematic global AI industry research of five major companies: Cisco, Dell, HPE, Lenovo, and Supermico.
IDC projects growth of non-x86 servers will moderate to 36.5% in 2026. See: IDC, “Servers market insights,” Oct. 21, 2025.
World Semiconductor Trade Statistics, “Global semiconductor market approaches $1T in 2026,” press release, Dec. 2, 2025; Yoana Cholteeva, “Global semiconductor market approaches $1T in 2026,” SDxCentral, Dec. 4, 2025; A (spring 2026) Deloitte study on the AI chip market initially estimated that AI chips in 2026 would be about $300B. Given the December 2025 upward revision in the global chip market of $175B by the World Semiconductor Trade Statistics (all of which was driven by AI demand, with weakness in non-AI markets), Deloitte now estimates that the AI chip market in 2026 will be about $500B.
Jarvis Labs, “NVIDIA H100 price guide 2026: GPU costs, cloud pricing & buy vs rent,” Jan. 8, 2026.
NVIDIA, “Cooling and airflow optimization”, accessed Jan. 14, 2026.
Paul Schell, “Top five AI server companies for data centers and enterprises”, ABI Research, Sept. 22, 2025.
Eric Huang and Jingyue Hsiao, “EMS watch: The year AI servers broke the EMS rankings,” DIGITIMES Asia, Jan. 22, 2026.
Duncan Stewart et al., “Why AI’s next phase will likely demand more computational power, not less,” Deloitte Insights, Nov. 18, 2025.
The 800V DC seems to be a promising solution, offering benefits such as efficiency, fewer conversions, optimal copper use, smaller size, and, importantly, low heat. To read further, see: Mathias Blake, Martin Hsu, Ivan Goldwasser, Harry Petty, and Jared Huntington, “NVIDIA 800 V HVDC architecture will power the next generation of AI factories,” NVIDIA Developer Technical Blog, May 20, 2025.
Ray Wang, “NVIDIA’s new Rubin CPX targets future of large-scale inference,” The Futurum Group, Sept. 18, 2025.
There are two different kinds of liquid cooling systems (direct-to-chip and full immersion) that multiple tech industry players—ranging from large existing companies providing end-to-end data center services to more specialized liquid cooling providers—offer in the market. To read further, see: Data Center Asia, “Understanding the different types of liquid cooling systems for data centers,” accessed Jan. 14, 2026.
Albert Hung, et al., “AI datacenters. Power and liquid cooling model update: Assessing PSU, liquid cooling, and electric grid impacts,” J.P. Morgan, Asia Pacific Equity Research, November 4, 2025. No link available.
Ibid.
Communication can be made between the various trays inside a rack (scale up), between the various racks in a data hall (scale out), and between different data centers (scale across). The fiber optics can be at very high speeds (up to 1.6 Terabit/second), sometimes using separate optical components, and sometimes using co-packaged optics (CPO) where the optical parts sit directly on top of the same silicon the AI chips do. See: Sharada Yeluri, “Co-packaged optics — a deep dive,” APNIC Blog, May 7, 2025.
Morningstar, “Generative AI’s rise: Impacts on networking for advisors,” Sept. 22, 2025.
Deloitte analysis based on multiple secondary sources including: Alfonso Maruccia, “Sandisk and SK Hynix partner on High Bandwidth Flash, promising up to 16× more memory for AI GPUs compared to HBM,” Techspot, Aug. 8, 2025; Trendforce News, “What comes after HBM? NAND-stacked HBF may power AI’s future,” Sept. 15, 2025.
Diana Kearns-Manolatos, “As cloud costs rise, hybrid solutions are redefining the path to scaling AI,” Deloitte Insights, Nov. 5, 2025.
Sascha Brodsky, “The hidden costs of AI: How generative models are reshaping corporate budgets,” IBM, accessed Jan. 30, 2026; Chris Thomas and Diana Kearns-Manolatos, “AI workloads are surging. What does that mean for computing?,” Deloitte Insights, Aug. 21, 2025.
Ashish Nadkarni, “AI infrastructure in 2025: Balancing datacenter and cloud investments,” IDC, February 2025.
Chris Thomas et al., “Is your organization’s infrastructure ready for the new hybrid cloud?” Deloitte Insights, June 30, 2025.
Duncan Stewart et al., “Telecoms tackle the generative AI data center market,” Deloitte Insights, Sept. 16, 2024; Doug Van Dyke et al., 2025 global telecommunications outlook, Deloitte Insights, Feb. 20, 2025.
Stewart et al., “Why AI’s next phase will likely demand more computational power.”
Mary Cunningham, “You’re living in a ‘K-shaped’ economy. Here’s how that affects you.,” CBS News, Nov. 5, 2025; Christopher Rugaber, “Here’s why everyone’s talking about a ‘K-shaped’ economy,” AP, Dec. 1, 2025; Mark Zandi, post on X.com, Sept. 16, 2025; IDC, “Global memory shortage crisis”; Robinson, “Budget smartphones will be hit hardest.”
The top 10% of US earners are now estimated to be driving nearly half of all consumer spending—and higher-income outlays are growing several times faster than lower-income demand. See: Cunningham, “You’re living in a ‘K-shaped’ economy.”
Steve Feinberg et al., “In the gen AI economy, consumers want innovation they can trust,” Deloitte Insights, Sept. 25, 2025.
In this paper, “lower income” is defined as average annual pre-tax household income below $65,000 (corresponding to below 40% of US income distribution), “middle income” as $65,000 to under $175,000 (40% to less than 80% of US income distribution), and “high income” as $175,000 and above (80% to 100% of US income distribution). These groups correspond to categories illustrated in Moody’s chief economist’s analysis of federal data. See: Zandi, post on X.com; PK, “Average, median, top 1%, and all United States household income percentiles,” DQYDJ, accessed Dec. 16, 2025.
For instance, 16% of lower-income respondents say they won’t be able to afford the devices their household needs and 21% expect they can’t afford more advanced products they want—challenges reported by only 5% and 4% of high-income respondents, respectively.
The income groups differ on how much price shapes tech purchasing decisions: Affordability is “very important” to 67% of lower-income consumers, 57% of middle-income, and 41% of high-income households.
Segments were created based on analysis of consumers’ responses to 16 survey questions that evaluate the perceived performance of their tech providers across innovation, intelligence, personalization, and core data responsibility dimensions like protection, transparency, and user control. Responses did not include the names of their tech providers.
link