Neocloud providers challenge a decade of hyperscaler control

Before the explosion of AI interest in 2023, the hyperscalers had enjoyed a nearly unimpeded growth trajectory since 2017, when the last of the telco cloud competitors exited the business. Verizon, IBM, Rackspace and a multitude of others all started their own cloud platforms in a bid to challenge Amazon Web Services (AWS) in the cloud infrastructure space but ultimately exited the space and focused on the periphery of the cloud market opportunity. Even the challenges posed by General Data Protection Regulation (GDPR) and other regulations were ultimately not enough to dislodge the major cloud providers, as the three leading U.S.-based firms still hold the largest market share, even in the heavily regulated European Union markets.
 

More than a decade after the market came into existence, the cloud infrastructure market remains a steady source of double-digit growth, with a total opportunity size of roughly $500 billion globally in 2025. Most of this market opportunity is claimed by AWS, Microsoft and Google Cloud, which together account for half of the total market. Although we expect the collective share of those three vendors to continue growing through 2029, neocloud providers will cause real disruption and limit the leading hyperscale cloud providers’ nearly unimpeded ability to expand. Although neocloud providers will not realistically capture leadership of the cloud infrastructure space, they will disrupt the hyperscaler market by capturing AI-related market growth, pressuring pricing for AI workloads, building a platform ecosystem around their services, forcing hyperscalers to partner with major neocloud providers, and directly targeting enterprise customers.

It is not just the revenue but also the growth that hyperscalers will miss

The most direct and visible impact of neocloud providers will be the revenue they generate. The neocloud market is estimated to have surpassed $25 billion in 2025, representing triple-digit growth year-to-year. That presents a significant opportunity and represents one of the fastest-growing segments of the overall cloud market. However, there is overlap, as many hyperscalers are also the largest neocloud customers, and the fact that this group of companies is capturing tens of billions of dollars and growing at a rapid pace is a complicating factor in the market. CoreWeave, for instance, earns nearly $2 billion in revenue each quarter, a figure that is roughly doubling year-to-year.

Neocloud providers will also pressure hyperscale pricing and margins

CoreWeave’s $2 billion in quarterly revenue is even more significant given the pricing advantages it offers its customers. Although the specifics vary based on a number of factors, in general, neocloud providers’ prices are 30% to 60% lower than what the major hyperscalers charge for the same services. That means CoreWeave is taking between $2.6 billion and $3.2 billion in market opportunity off the table from hyperscalers. The total revenue impact is even more severe after accounting for the pricing pressure those hyperscalers are forced to grapple with while trying to minimize the disparity between their AI service prices and neocloud offerings.
 

Hyperscalers’ expenses and margins will also reflect the impact of neocloud providers, as their operating models vary significantly, as shown in Figure 1. CoreWeave, for instance, is operating at basically a break-even profit level, investing internally, largely through R&D and infrastructure build-outs. Neoclouds’ increased pricing and investment pressures will also impact AWS’ double-digit operating margins.
 

Figure 1: 2025 Operating Expenses as a Percentage of Revenue for CoreWeave and Amazon Web Services (Source: TBR)

Building out a platform will solidify neocloud competitive positioning

While selling access to the raw GPU capacity remains the primary way in which neoclouds are disrupting the hyperscale landscape, expanding into the platform layer of services will have a sustained impact. As GPU supply expands and competition intensifies, platforms have become critical for differentiation, customer retention and higher-margin services. This is a strategy straight out of the cloud hyperscaler playbook, as AWS, Microsoft and Google have entrenched themselves with customers through additional development, integration, marketplace and data services on top of their core infrastructure capabilities. Neoclouds, by layering orchestration, AI tooling and developer environments on top of specialized infrastructure, aim to move up the value chain and compete more directly with hyperscaler AI environments.
 

CoreWeave provides one of the clearest examples of this platform strategy. The company now positions its offering as a purpose-built AI cloud platform that combines high-performance infrastructure with intelligent software tools. The platform integrates services such as managed Kubernetes environments, GPU-native scheduling systems and AI storage layers designed to support large distributed training workloads. For example, the CoreWeave Kubernetes Service (CKS) and Slurm-on-Kubernetes orchestration stack allow customers to efficiently schedule massive GPU jobs and maximize cluster utilization.
 

Beyond orchestration, CoreWeave has expanded into tools that support the broader AI life cycle, including training infrastructure, inference deployment and integrations with AI development ecosystems. The company’s platform is designed to enable developers to build, train and serve AI models within a single environment, rather than relying on fragmented tooling across multiple providers. These additional services target some of the fastest-growing addressable markets for hyperscalers, large-scale AI training and inference. The impact becomes more distinct in the long term, as customer adoption of platforms is quite sticky, preserving neocloud advantages even as GPU scarcity and price-to-performance advantages potentially fade over time.

Hyperscalers are forced not only to compete with neoclouds but also to partner and purchase from them

Although there is a competitive element between hyperscalers and neoclouds, both groups also rely on each other to capitalize on the AI market opportunity. This reflects the explosive demand for AI compute and the limits of hyperscalers’ ability to build capacity quickly enough to meet that demand. Neocloud providers such as CoreWeave have built their businesses around specialized AI infrastructure and have an inherent advantage in the space due to their unique relationships with NVIDIA, which allow preferential access to supply-constrained GPUs. As demand for generative AI accelerates, these specialized environments have become valuable sources of additional compute capacity — even for the largest cloud providers, leading to a paradoxical relationship between hyperscalers and neoclouds.
 

On one hand, hyperscalers must compete with neoclouds for AI customers, particularly startups and AI labs seeking large GPU clusters at competitive prices. On the other hand, hyperscalers also benefit from accessing the infrastructure that neoclouds have rapidly built. In some cases, hyperscalers purchase compute capacity from neocloud providers to supplement their own data center supply while their internal AI infrastructure continues to scale.
 

This hybrid competitive and cooperative dynamic reflects a broader shift in the cloud market. AI infrastructure demand has grown so quickly that no single provider can fully control the supply of compute resources. As a result, the emerging AI cloud ecosystem is becoming more interconnected, with hyperscalers, neoclouds and infrastructure providers operating within a complex web of competition and collaboration. In the near term, this dynamic is likely to persist, limiting the aggressiveness with which neoclouds and hyperscalers compete.

Expanding to enterprise customers is the next stage of neocloud diversification

Although platform services and capabilities represent neoclouds’ functional diversification, enterprise customers represent diversification efforts within the neocloud customer base. The supply-constrained environment and tremendous amounts of investment made AI startups and hyperscalers lucrative early customers for neocloud providers. That concentrated customer base has created a significant market very quickly, but expansion and diversification represent the next phases in the market’s evolution. Given the areas of competition with hyperscalers, it is particularly important to expand the customer base to large enterprises that build and run their own AI models. To expand into enterprise customer accounts, neocloud providers are combining specialized AI infrastructure with enterprise-grade platforms and long-term capacity agreements that appeal to organizations deploying large-scale AI workloads.
 

For instance, in addition to providing GPU-dense infrastructure optimized for AI training and inference and large clusters of NVIDIA accelerators connected through high-bandwidth networking, CoreWeave is also targeting enterprise customers by building a platform around AI workload management and deployment. The company’s managed Kubernetes environments and GPU-aware scheduling capabilities allow enterprises to run large distributed training workloads more efficiently. Integrations with AI development tools and machine learning frameworks enable customers to build, train and deploy models within a unified environment rather than assembling multiple infrastructure and software layers independently. CoreWeave has also pursued enterprise adoption through large, multiyear compute agreements that give customers guaranteed GPU capacity while ensuring predictable revenue streams for the company. Although neoclouds’ efforts to diversify their customer bases are still new, the trend will eventually lead to greater stability for the providers as the AI and GPU markets mature.

On Jan. 22, 2026, Fujitsu Americas hosted more than a dozen industry analysts in Toronto for a day of briefings on the company’s progress, specifically in the Americas. Speakers included Asif Poonja, EVP, corporate executive officer & CEO, Americas Region; John Slaytor, VP, head of Strategy & Uvance; Durga Kota, CTO, head of Data x AI, Fujitsu North Americas; Nicholas Lee, executive director & head of Fujitsu Intelligence; Sudhir Nair, CEO & co-head, managing partner, Uvance Wayfinders Americas; and Fleur Copping, VP, Strategic Alliances Unit, Regions. The following day, TBR recorded an episode of TBR Talks with Poonja in Fujitsu’s Toronto office, which went live March 6 on all streaming platforms. This special report highlights the information shared during analyst day, the TBR Talks conversation with Poonja, and TBR’s ongoing analysis of Fujitsu, both in the Americas and globally.

Fujitsu’s technology strategy

Artificial intelligence was emphasized during Kota’s discussion of the company’s technology strategy. He noted that Fujitsu’s top priority remains developing new intellectual property, with an emphasis on cocreating with clients. Fujitsu’s technology development teams also help accelerate the company’s overall portfolio transformation.
 
While this is inherently more of an internal strategy and less client-driven, Kota stressed that clients will eventually benefit from Fujitsu’s innovation, based on the company’s impressive track record to date. Kota noted that the company intends to be an intelligence integrator, bringing together all the technology in alignment with client needs.
 

In TBR’s view, the overall technology strategy — innovate with clients, invest in R&D and advancing technology, and bring it all together in a consistent manner — is not groundbreaking, but it fits well with Fujitsu’s culture, history and market position. Fujitsu has been a reliable innovator and has robust experience translating lab work into technology used by clients. In a market swamped with AI and tech hype, a sensible, practical tech strategy stands out for being reasonable, straightforward and executable.
 
Kota also commented on the well-known reality that clients have lots of data but do not know how to use it (noted in every edition of TBR’s Digital Transformation: Voice of the Customer Report since 2019). Fujitsu tackles the problems in part by examining different client personas:

• Chief operating officers want to take advantage of data, so Fujitsu brings operational efficiency and optimization.
• CTOs spend lots of money on technology but need to modernize, maintain and manage. Conveniently, CTOs can outsource all of that to Fujitsu.
  CFOs want to use AI to make changes to the business through scenarios, ontology-based decisions and automation, all of which are increasingly in Fujitsu’s wheelhouse, especially with the evolution of Uvance and Wayfinders.

In Kota’s and other Fujitsu leaders’ telling, the company’s differentiation comes from explainability (around AI), multiagent orchestration and governance. Fujitsu’s approach is to solve the biggest business problem that can feasibly be solved and then use the savings from that to fund work on smaller problems. By deploying at the client site, deepening relationships and moving faster, Fujitsu can sell clients on the target 5x savings, equivalent to 5x the price of Fujitsu services and solutions.
 
In TBR’s view, Fujitsu’s approach, with an emphasis on understanding personas and targeting savings, may not be markedly different from peers’, but the company remains focused on clients and outcomes, rather than internal metrics of IT services greatness. In TBR’s research, clients care about what you can do for them, not how you are different from peers. Fujitsu has the right focus and now needs to execute.

Uvance and Wayfinders

Fujitsu continues to develop its stories around Uvance and Wayfinders, including more tightly focused messaging that shows both differentiation from peers and a clear value proposition for Fujitsu’s clients and technology partners. Given the inherent challenges in repositioning Fujitsu in the market, TBR believes the company’s emphasis on developing the Uvance and Wayfinder stories is critical during 2026.
 
Slaytor discussed Fujitsu’s Uvance strategy, and Nair explained how Wayfinders will fit into Fujitsu’s shifting market position, noting that Wayfinders remains a global organization that works locally but is managed globally. Nair said Wayfinders brings global reach and an “end-to-end” understanding of Fujitsu’s solutions and services. Many of the Wayfinder professionals have been recruited directly from IT leadership roles within enterprises, bringing deep industry experience, or from leading consultancies.
 
Wayfinders serve as the “linchpin,” according to Nair, and benefit from the “tremendous technology, IP, capabilities, solutions and credentials” of Fujitsu as a whole. While leaning into an engineering mindset that remains foundational to Fujitsu, Wayfinder consultants can “broaden the conversation” to help clients “get to transformation.”
 
In TBR’s view, focusing on manufacturing clients, hiring directly from industry, and positioning Wayfinders as part of, but uniquely separated from, Fujitsu all makes sense … for now.
 
Building a consultancy requires continually solving two challenges: people and permission. Technology-centric IT services companies that have sought to build management consultancies in-house have typically struggled with a few of the following challenges, all of which tie back to people and permission: 1) scale; 2) brand, both internally and externally; 3) talent acquisition and retention; 4) over-indexing on lessons learned from a small sample of early engagements; and 5) culture, which changes slowly, if at all.

Can Fujitsu succeed in 2026?

In terms of success in 2026, a few factors weigh heavily in Fujitsu’s favor, including leadership commitment, a tightly proscribed value proposition, and excellent technology chops at a time when every consulting engagement is a technology engagement.
 
Perhaps the biggest challenge will be Fujitsu’s ability to convince clients and technology partners that the company’s consulting capabilities are on par with competitors’ and of the quality that everyone has come to expect from Fujitsu. The Wayfinders do not need to convince “the market” that they are bringing value; they just need to convince the right selection of Fujitsu’s clients — and Fujitsu’s leadership. 2026 should be a banner year for Uvance and Wayfinders.

Alliances

Fujitsu has become something of a TBR favorite when discussing technology alliances strategies. The company’s relationship with ServiceNow is a case study in aligning sales teams, and Fujitsu has consistently demonstrated leadership commitment to alliances across the ecosystem.
 
In Toronto, Copping continued to build on an impressive alliances story by detailing Fujitsu’s relationships with the top six strategic partners — Microsoft, Amazon Web Services (AWS), Oracle, SAP, ServiceNow and Salesforce — as well as developing alliances, such as Fujitsu’s investment in partnering more closely with Dynatrace. Copping explained precisely what Fujitsu looks for with its technology partners (echoing sentiments shared with TBR in early 2025): invest both financial resources and technology skills; help Fujitsu de-risk delivery of a partner’s technology; and provide access to innovation and the newest technology developments.
 
Illustrating execution of Fujitsu’s alliances strategy, Copping described two Oracle engagements, which both started as a client faced a messy, complex IT environment, saddled with technology that had outlived its usefulness, as well as a telecom engagement that showed how technology can propel and paralyze decision making. The three examples underscored a sentiment TBR has heard repeatedly since mid-2025: “I’m not sure every customer is ready for [the AI solutions] the tech companies are pushing.”
 
In wrapping up the alliance discussion and cementing Fujitsu’s place as a leading player in ecosystem management, Copping noted that Fujitsu’s 1,400 cybersecurity professionals were being cross-trained on ServiceNow and Fujitsu’s ServiceNow-related capabilities. Rather than train ServiceNow experts on cyber, Fujitsu’s focus has been to broaden ServiceNow skills — most critically, understanding — across an existing and profitable practice. TBR is frequently asked, “What separates the best ecosystem players from their peers?” This ServiceNow approach will now be part of the answer.

What’s next?

TBR will incorporate the numbers, data, details, strategies and case studies Fujitsu shared during its Toronto presentations into ongoing analysis, including a deeper examination of Uvance, the Wayfinders and Fujitsu’s AI differentiation.
 
One element that struck TBR was the intentional and stark difference between the company’s U.S. and Canada practices. This was evident during the event and made clearer by physically being in the Toronto office, and is also an aspect of Fujitsu’s overall North Americas strategy. Although Fujitsu is one company unquestionably working together and supported as a single region by the company’s global parent in Japan, the U.S. business is Manufacturing and the Canada business is Public Sector, not exclusively, but Fujitsu has — wisely, in TBR’s view — played to its strengths in Canada and Japan and avoided the temptation to make managing easier by harmonizing target client sets, go-to-market motions and marketing. Even as the Uvance and Wayfinder efforts catalyze change across the overall company, TBR anticipates the well-led and distinctly focused Canada and U.S. practices will continue to differentiate Fujitsu in the Americas.

AI PC Ambitions Face an Unforgiving Reality of Memory Constraints and Budget Pressure

For the PC industry, 2025 was the year that the end of Windows 10 support would drive a massive PC refresh cycle. As part of this refresh, AI PCs, devices with neural processing units (NPUs) designed to execute AI and machine learning tasks, were expected to rise in popularity as businesses prepared their workforce for a new era of AI-infused productivity. The PC refresh cycle of 2025 did materialize to some extent, with the market growing about 5.3% year-to-year according to TBR’s estimates, but sellers ran into a snag in the commercial PC market: Many buyers were unwilling to pay 10% to 20% more for an AI PC that could future-proof their PC fleet but offer minimal added value today. This sentiment has driven ecosystem players such as Intel and Microsoft to invest in making the AI PC more compelling to and affordable for the commercial PC market. The beginning of 2026 ushers in the latest generation of AI PC processors from Intel, AMD and Qualcomm that is poised to attract customers with performance, efficiency and security enhancements. In many ways, 2026 is a critical year for the success of the AI PC due to the significant investment in the platforms that have been years in the making.
 
When vendors made these long-term investments in next-generation AI PCs, no one knew that these new models would launch just as insatiable demand in the AI server market caused an unprecedented shortage of memory. The shortage, which in TBR’s opinion could easily last for two to three years, has already caused significant price hikes in the PC market — with more on the way — and will result in longer lead times and lower availability. As such, PC market leaders’ desire to usher in a new era of more expensive AI computing now sits in direct odds with unavoidable component price increases and tight customer budgets.
 

Pressure to refresh drove 2025 growth, with the intrinsic value of AI PCs yet to be defined

In 2025 AI PCs capable of meeting Microsoft’s Copilot+ requirements began to hit the market in earnest, following their initial debut in the summer of 2024. On paper, the timing looked favorable: A Windows 11 refresh cycle was underway, and AI PCs were positioned to ride that momentum and steadily grow share within the broader commercial PC market. In practice, however, the refresh materialized unevenly and ultimately fell short of expectations. Although the market experienced solid growth, expanding an estimated 5.3% year-to-year in 2025, multiple structural and behavioral hurdles slowed adoption of AI PCs, preventing vendors from translating technical readiness into meaningful volume growth.
 
Pricing pressures emerged as a central inhibitor. Tariffs complicated OEM pricing strategies, and the incremental cost of NPU-enabled processors introduced a noticeable price increase that many commercial buyers were unprepared to absorb. This sticker shock was exacerbated by the lack of a compelling near-term value proposition for AI PCs. Vendors leaned heavily on the concept of future-proofing device fleets, a message that resonated poorly with budget-constrained IT buyers prioritizing immediate, tangible ROI. At the same time, commercial purchasing behavior has yet to normalize post-COVID. Some organizations are still digesting inventory acquired during earlier buying surges while others have permanently shifted from predictable refresh cycles to more opportunistic, as-needed purchases.
 
In response, OEMs and key ecosystem partners — notably Microsoft, Intel and AMD — adjusted their go-to-market strategies to reduce friction and keep AI PCs moving into commercial environments. OEMs refined system configurations where possible to lower bill-of-materials costs without undermining Copilot+ eligibility. Silicon providers increased front-end discounting, rebates and comarketing investments to narrow the price gap between traditional PCs and AI-capable systems. Together, these efforts drove down pricing of AI PCs in the commercial space and helped some customers make the transition to NPU-enabled silicon during their device refreshes.

Ecosystem investments years in the making underpin the future success of AI PCs

Despite increasing competition from AMD and Qualcomm, Intel remains the PC silicon market share leader due in large part to the robustness of the company’s partner and incentive programs. PC OEMs and the broader industry supplying PCs rely heavily on Intel to influence market direction and demand. Understanding that the AI PC market is somewhat nebulous due to ill-defined use cases, Intel has invested significantly to make its latest chips — Core Ultra Series 3 — a success, leaning into its heritage as an integrated device manufacturer and shifting manufacturing of the compute tile on the new system on a chip (SoC) in-house rather than outsourcing to Taiwan Semiconductor Manufacturing Co. as it did with the Core Ultra Series 2. Although the investment to bring 18A capacity online has been massive, in the long run Intel could benefit from a cost of sales perspective with the manufacturing based in the U.S. and improved yields as the process matures. In contrast, as fabless companies, AMD and Qualcomm rely solely on third-party manufacturing, which presents fewer risks but could prove more costly in the long run.
 
Silicon vendors’ investments in driving AI PC adoption have looked similar from company to company, with AMD, Intel and Qualcomm providing developer resources such as samples, guidance and reference applications to drive NPU-enabled application development. However, although AMD’s and Qualcomm’s resources are limited to software developer kits and other software stacks directly targeting the NPU, Intel’s play also leans into its reputation as a market maker with its AI PC Acceleration Program, incentivizing collaboration between ISVs and independent hardware vendors (IHVs) to create features that leverage the NPU. Additionally, with the engineering staff available through Intel Open Labs serving as a differentiator, Intel arguably provides more design resources, sales support and go-to-market funding than AMD and Qualcomm, echoing the company’s long-running strategy that has positioned it atop the Windows PC silicon market. Similar to what silicon vendors are providing, Microsoft has launched tools that will help ISVs refine their Windows-based apps to make the best use of CPU, GPU and NPU.
 
Qualcomm, more so than Intel and AMD, needs to prove itself in the commercial market and demonstrate not only that Arm can outperform x86 in commercial uses but also that the platform is a trustworthy investment. However, much of this weight falls on Microsoft’s shoulders as Windows on Arm is inherently its own initiative. Perhaps more important than its advances in performance in TOPS (trillions of operations per second), Qualcomm’s newly launched Snapdragon X2 Plus platform offers a competitive answer to Intel’s vPro platform with Snapdragon Guardian, which allows for advanced remote management and security features. However, Intel vPro’s Active Management Technology, which allows for out-of-band remote management, remains core to the company’s value proposition, is still unmatched by Qualcomm and AMD, and is a key reason noncloud-first organizations tend to choose Intel over its competition.

2025 Windows AI PC SoC Revenue by Covered Vendor [Intel, AMD, Qualcomm] (Source: TBR)

Massive scale AI infrastructure deployments are driving unprecedented fluctuations in memory costs, negatively impacting supply for the PC market

The PC market has weathered many component shortages that have had short-term impacts on product availability and pricing. In recent history, shortages have been caused by factors ranging from pandemic-induced buying changes and natural disasters to manufacturing industry consolidation and delays in next-generation technology rollouts. However, the current memory component shortage is unique due to its ties to structural AI demand versus a temporary shock to the market. Over the past two years, demand for AI infrastructure has shifted the slow-growing server market into hypergrowth, with TBR estimating that server spend nearly doubled from 2023 to 2025. Not only has the demand risen dramatically, but the AI server silicon, whether GPUs or AI ASICs, also requires significant amounts of high-bandwidth memory (HBM), unlike their traditional server counterparts.
 
As such, hyperscalers and neoclouds are absorbing a disproportionate share of DRAM and HBM as they scale AI training and inference infrastructure, and the same suppliers sell both HBM and conventional DRAM. As these memory suppliers prioritize higher-margin HBM, commodity DRAM output growth tightens as investments in new capacity are allocated more heavily toward supporting AI infrastructure rather than client devices.
 
This trend of elevated server demand combined with disproportionate consumption of high-performance memory will not abate anytime soon. OEMs’ total server order backlogs remain in the billions exiting 2025, and during AMD’s 4Q25 earnings call CEO Lisa Su reported the company expanded manufacturing capacity for server CPUs, confirming that component suppliers continue to shift their manufacturing lines in favor of servers rather than PCs.
 
As a result, PC DRAM supply has not grown enough to support increasing demand. The DRAM found in thin-and-light PCs is the same DRAM found in most smartphones, and the replaceable DRAM typically found in desktops and workstations is also used in traditional enterprise servers. Due to this supply-and-demand imbalance, prices for PC DRAM have skyrocketed and are likely to remain elevated through at least 2026 and into 2027. Already, contract prices for the DRAM found in desktops and workstations have increased between 10% and 30% year-to-year going into 2026 with the DRAM found in thin-and-light PCs increasing at an estimated rate of between 20% and 40%.
 

OEM Server Revenue in $USD Billions for 2020 through 2025 (Source: TBR)

Memory costs and longer lead times put AI PC momentum at risk

TBR expects the memory shortage will impact commercial PC buyers in three ways: pricing, availability and product assortment. Channel partners in contact with TBR have reported PC OEM prices have increased twice since the beginning of 2026, with more price hikes planned. Although price increases to date have been estimated between 10% and 20%, a partner noted a customer whose specific PC configuration price had increased by 40% since December 2025. Channel partners have also noted that delivery times have already become longer, meaning customers will have to accept longer lead times going forward.
 
These dynamics put AI PC models with higher-priced silicon in a precarious position. Customers with finite spend are already facing significant price increases on memory that offers no additional value. Many customers will face a choice of further delaying PC refresh, reallocating budget from other IT projects, or seeking watered-down configurations to reduce costs.
 
OEMs will need to maintain tight cost controls as much as possible through supply chain management and focus on product assortment. For example, OEMs will likely push more models with 16GB of RAM, which is the minimum requirement for Copilot+ PCs, over those with 32GB or 64GB. Ultimately, with longer lead times and a tight product assortment a likely outcome, TBR expects OEMs will focus on configurations and segments where they can maintain margins and best serve the customers that are willing to pay.
 
The cost burden will not fall solely on the OEMs. With all three AI PC silicon vendors launching new chips at CES in January, these vendors are motivated to make the new platforms a success. OEMs and silicon vendors already had to adjust pricing on NPU-enabled silicon in 2025 to reach a point where customers were willing to upgrade. TBR expects OEMs will pressure Intel, AMD and Qualcomm to provide additional discounting to ensure the success of the latest generation of chips.

AI PC Revenue by Big Three Windows OEMs [Dell Technologies, HP Inc., Lenovo] for 2025 and 2026 (Source: TBR)

2026 outlook: Increasing prices will stifle overall market growth and slow AI PC mix shift

Despite significant pricing impacts, TBR expects the PC market will grow in 2026. In fact, the early part of the year may perform better than expected if customers pull in buying to avoid additional price increases coming down the road.
 
Although the PC ecosystem has been investing in making NPU-enabled silicon more useful, the use cases for most AI PC buyers remain thin and focused on future innovation. The AI PC mix will continue to increase, but TBR believes the growth rate will be lower than we predicted in 2025. The increase in mix will be based more on the availability of chips and manufacturers’ emphasis on AI PCs than on organic customer demand.
 
Although TBR estimates the overall value of the PC market will grow 1.7% year-to-year in 2026, the increase will be driven by rising selling prices, not shipments, which are expected to decline as many buyers delay PC refreshes unless absolutely necessary. Meanwhile, TBR forecasts the overall value of the AI PC market will expand 33.4% in 2026 due in large part to a relatively small compare and PC OEMs’ (excluding Apple) growing AI PC revenue mix, which TBR predicts will increase from 23.9% in 2025 to 37.9% in 2026, a difference of 1,400 basis points. Although this mix shift is pronounced, TBR firmly believes that had PC DRAM prices remained the same, the increase would be more robust.
 

TBR’s AI PC Market Forecast for 2025 through 2030 (Source: TBR)

Conclusion

Rising demand for the types of DRAM found in PCs, in combination with memory suppliers focusing production capacity build-out on supporting massive-scale AI infrastructure deployments, has led to a supply-and-demand imbalance in the market, causing memory manufacturers and suppliers to rapidly increase PC DRAM prices. With margins already thin, PC OEMs are passing their rising build costs on to buyers in an attempt to protect profitability. AI PC prices were already higher than traditional PC prices, and with AI PCs typically configured with more memory, on average, AI PC prices are poised to continue increasing more than traditional PC prices, making the future-proofing-led sales proposition less compelling. Not only will these rising prices result in slower-than-anticipated adoption of AI PCs, but they will also cause more organizations to delay their PC refreshes. However, once PC DRAM prices stabilize and begin to fall, pent-up demand from delayed refreshes could yield a prosperous cycle for PC OEMs and silicon vendors alike.
 
Fortunately, although the rate at which memory prices are increasing is unprecedented, PC OEMs have weathered component shortages in the past. While increasing PC prices will slow the market’s transition to AI PCs in the near term, the proliferation of the AI PC over the next five years is almost guaranteed due to the alignment of ecosystem players, including PC silicon vendors, PC OEMs and Microsoft, which owns the operating system. As such, increasing memory costs and PC prices are less of a roadblock to AI PC adoption and more a speed bump impacting the momentum of adoption.