Why businesses should be worried about RAM prices

RAM prices have shot up thanks to increased demand by data center operators for AI workloads

An illustration of a rising price arrow, going from blue to red in color, set against a dark green outline of a RAM chip against a striped bright green background.
(Image credit: Getty Images)

There is a structural shift underway across the whole memory market. Random access memory (RAM) is at the heart of modern computing, used across swathes of devices – from ultraportable laptops to tower desktops and everything in between, including smartphones and tablets.

But prices are surging due to an imbalance between supply and demand: a relatively constricted supply has come to loggerheads with demand by data center operators that require specialized memory components for use in AI applications.

Far from being abstract, this will have material consequences for every single business through the prism of hardware provision, with the effect varying depending on the nature of your organization. But what's causing the crisis, and how long is it likely to last?

Why are memory prices surging?

This unprecedented memory chip shortage was first spotted toward the end of 2025. Graphics prepared by PC Part Picker illustrate the severity of the price rises, with RAM roughly doubling in price and the DDR5-6000 2x32GB memory kit tripling. IDC then published a detailed blog post in December that amounted to a serious warning.

"For consumers and enterprises alike, this signals the end of an era of cheap, abundant memory and storage, at least in the medium term," IDC analysis concluded in their report. "The year 2026 is shaping up to be one in which technology becomes more expensive, driven by supply constraints rather than demand growth."

The analyst firm highlighted a handful of factors at the heart of the price surge. Chief among them was the rapid expansion of AI infrastructure and workloads, causing manufacturers to reallocate capacity to building components specifically designed for AI, rather than conventional devices. Memory manufacturers, he wrote, have shifted production to high-bandwidth memory (HBM) and high-capacity DDR5 RAM, rather than components typically used in laptops or smartphones, driving up prices across the board.

"Why manufacture low-margin wafers when you can make high-margin ones?" CEO of NPI Financial, Jon Winsett, tells ITPro. "The big manufacturers have shifted by over three times their capacity to high-bandwidth memory production, which commands significantly higher profit margins than commodity DRAM. Another factor is stockpiling by hardware manufacturers to hedge against inventory depletion."

Although comparisons with the COVID-19-induced chip crisis may be attractive, Winsett says that it was a "different beast" driven by a cascade of impacts to the supply chain. Here, the outcomes are more "intentional" in nature. "Memory manufacturers are opting to make and sell chips to data centers as they make nearly five times the profit that way. There’s little incentive to provide the chips needed for enterprise PCs/servers/storage when they can make that much more money," he explains.

Gui Soubihe, founder of digital infrastructure provider Latitude.sh and now head of compute at Megaport, agrees. He says external disruptions sparked the pandemic shortage, whereas this current shortage is a "strategic choice by manufacturers prioritizing high-margin AI memory over commodity DRAM".

"With three vendors controlling the market and actively exiting consumer lines — Micron is shutting down its Crucial brand [for example] — we can say with confidence that shortages won’t be resolved in the near future," he adds.

The downstream impact of these pricing increases will find their way into every aspect of IT in 2026

Jon Winsett, CEO at NPI Financial

What are the implications for businesses moving forward?

Soubihe notes that RAM prices have surged by as much as six-fold in the last few months with availability dwindling. Companies that have not secured IT equipment they need should be ready to "pay up or get in the back of a very long procurement line". It's a stark and rather dire forecast, with IDC's report describing the end of an era in terms of cheap memory components. "What began as an AI infrastructure boom," they wrote, "has rippled outward, with tightening memory supply, inflating prices, and reshaping product and pricing strategies across both consumer and enterprise devices".

Winsett tells ITPro that the cycle will last longer and hit IT budgets much harder than many may anticipate. That is, on top of finding available hardware in the first place, and dealing with longer delivery times.

"Companies should accelerate refresh decisions where possible, and lock in pricing and configurations early. It’s important to push for extended quote validity as well," he says. "We’re seeing some hardware OEMs reprice mid-deal. Standard builds of devices should be reevaluated to avoid unnecessary exposure to premium memory tiers. Lastly, and more important than ever, is to benchmark hardware pricing. Assume you’re being quoted higher than fair market value. That goes for software and cloud too. The downstream impact of these pricing increases will find their way into every aspect of IT in 2026."

While enterprises and consumers are likely to face much higher costs, IDC highlights the potential impact on the original equipment manufacturers (OEMs) at the heart of the supply chain. This cycle is likely to have a serious negative impact on smartphone and PC shipments, while also hitting OEMs' ongoing 'AI PC' sales narrative. While modern devices need more RAM to power local large language models (LLMs), for example, higher prices could put a major dent in the manufacturer drives to convince businesses to roll out AI-ready devices across the breadth of their enterprise. This could further hamper the purported productivity gains working professionals gain from mainstream AI.

Unfortunately, we only seem to be at the beginning of this new cycle. Winsett says there's a chance it could be resolved by the end of the year, but realistically, we're looking at 2027 or even 2028. "The AI bubble would have to pop to see major revisions in demand, as increases in supply will take at least one to two years to have a meaningful impact on pricing," he explains. "But unless we see new memory market entrants or a collapse in AI demand, we could see 10-20%-plus quarterly price increases in prices well into 2027 — or longer."

There is, sadly, seemingly little that the majority of businesses can do to mitigate these additional costs when it comes to, say, their hardware refresh cycles. That is, unless they prioritize fast-tracking purchases or even negotiating extended deals with suppliers where prices are fixed. All that's left for businesses to do is ride out the wave and hope that production capacity expands as soon as possible.

Keumars Afifi-Sabet
Contributor

Keumars Afifi-Sabet is a writer and editor that specialises in public sector, cyber security, and cloud computing. He first joined ITPro as a staff writer in April 2018 and eventually became its Features Editor. Although a regular contributor to other tech sites in the past, these days you will find Keumars on LiveScience, where he runs its Technology section.