An escalating shortage of memory chips has reached the crisis stage, according to analysts, and will significantly impact smartphone and PC markets in 2026, as well as edge applications like smart glasses, wearables and industrial machines that rely on a mix of memory, compute, sensors and other components.
“The situation is both severe and complex, and manufacturers’ top executive teams are treating it as a critical priority, given its potential to materially affect profitability and, for some brands, threaten the viability of their businesses,” wrote Francisco Jeronimo, vice president for client devices at IDC.
Jeronimo’s commentary and concerns, shared with Fierce and IDC clients, is being echoed by sensors and electronics makers. Three executives in the sensors industry who asked not to be identified told Fierce in the past week that they can still plan prototypes of circuit boards and other components with existing memory stockpiles, but the longer-term outlook for production of their components is less than optimistic given the memory shortage and potential for much higher prices as the shortage continues. They each said they are following the memory shortage impact on the broader electronics industry closely.
The concern over memory will be dire for OEMs that make PCs, smartphones and products sold into industrial and consumer markets, since they keep regular production schedules that require a continuous supply.
In smartphones, low end suppliers like TCL, Xiaomi, Oppo and Huawei, “are likely to suffer significantly,” Jeronimo said. Increased memory costs will hit their margins and they will pass on part or all of the cost to end users.
As a result, smartphone average selling prices will potentially rise up to 8% and the overall smartphone market could contract by as much as 5.2%, IDC said.
For PCs, the memory shortage “creates a perfect storm,” he said. Lenovo, Dell, HP, Acer and ASUS have already warned clients of tougher conditions ahead, confirming up to 20% hikes in prices.
Memory shortages will also affect white box PC vendors, many of whom are used in edge and industrial products that rely heavily, in recent trends, on sensors, memory, AI software and compute integrated circuits.
“White box vendors…will bear the greatest burden of the shortage and that will include DIY systems oftentimes built by gamers,” Jeronimo said.
He even asserted that the memory shortage “threatens to derail the industry’s growth narrative around the AI PC.” Overall, IDC predicted up to an 8% hike in average PC selling prices, with up to an 8.9% contraction in the PC overall market, depending on how long current supply constraints persist into 2026.
“For consumers and enterprises alike, this [shortage] signals the end of an era of cheap, abundant memory and storage, at least in the medium term,” Jeronimo added. “The year 2026 is shaping up to be one in which technology becomes more expensive, driven by supply constraints rather than demand growth.”
IDC’s warning is broadly shared by many electronics technology analysts. Jack Gold, president and principal analyst at J. Gold Associates, put it this way in comments to Fierce: “The impact of the shortage will be felt throughout the industry…I suspect that the shortages will have a pricing and supply impact on edge and even smaller devices. I don’t think it’s time to panic, but the shortages are real and will have an effect.”
Microchip USA, a distributor of memory chips, wrote as early as September, “The global memory shortage is real and accelerating” with HBM memory demand surging due to AI and data-center expansion.
TrendForce has reported prices in some memory segments have more than doubled since February. These segments include flash chips used in USB drives and smartphones, all the way to high-bandwidth memory for AI chips in data centers. In one example, A 1 Tb TLC NAND chip that cost about $4.80 in July jumped to $10.70 by November. A 16Gb DDR5 chip rose from $6.84 in September to $27.20 in December—a nearly 300% in three months, according to TokenRing AI, in a news report from WRAL News.
What’s the cause of the memory shortage?
Every analyst blames the memory shortage on the AI frenzy, as memory makers shift gears to supply high bandwidth memory and other memory and storage products to data centers being built or expanded to handle AI demands.
“The memory shortage has now graduated from a component level concern to a macroeconomic risk,” wrote Sanchit Vir Gogia, CEO of Greyhound Research, in recent weeks. The AI buildout “is colliding with a supply chain that cannot meet its physical requirements.”
“As AI draws the big ticket memory suppliers away from the PC and phone markets, it creates additional supply shortages there,” Gold said, adding he suspects the same will apply to small devices. “Why build $5 to $10 chips for consumer products when you can build $50 to $100 chips for AI?”
Micron recently said it would no longer produce PC and consumer-based memory under its Crucial brand to concentrate on high value memory for AI and data centers. That alone will have an impact on supply because Micron is a major memory producer along with SK Hynix and Samsung.
The memory chip shortage is “unprecedented” and will have knock-on impacts for device makers and end users well into 2027, Jeronimo added. He predicted “a potentially permanent, strategic reallocation of the world’s silicon wafer capacity…
“The voracious demand for HBM by hyperscalers, such as Microsoft, Google, Meta and Amazon, has forced the three biggest memory manufacturers—Samsung, SK Hynix and Micron—to pivot their cleanroom space and capital expenditure towards higher margin enterprise-grade components. This is a zero-sum game: every wafer allocated to an HBM stack for an Nvidia GPU is a wafer denied to the LPDDR5X module of a mid-range smartphone or the SSD of a consumer laptop.”