Micron recently boosted its outlook for server shipments.
Micron sells both HBM memory for AI chips and standard server DRAM, both of which are in short supply.
Intel is struggling to keep up with demand for server CPUs, although capacity constraints are expected to ease later in 2026.
10 stocks we like better than Micron Technology ›
The growing demand for servers is surprising even the most optimistic AI hardware providers. In December, Micron(name: mu) strengthened its outlook for the growth of the server unit in 2025 from 10% to a high percentage of teenagers. The update was notable both for the scale of the increase and for the fact that it came so late in the year. Looking ahead to 2026, Micron sees strong growth in demand continuing.
The value of shipped servers is also increasing, driven by the shift to AI servers featuring expensive GPUs. IDC predicts an 80% increase in global server spending in 2025, and another 24.3% increase in 2026.
Image source: Getty Images.
This environment benefits Micron, but it also benefits Intel(NASDAQ: INTC). Intel leads the market in data center CPUs, and although it has been dropping market share AMD for years, the CPU giant is now struggling to keep up with demand as hyperscalers ramp up orders.
Memory chips largely act like a commodity, so in times of insufficient supply, prices can rise dramatically. That’s exactly what’s happening right now, and increased demand for servers will push prices even higher.
Micron, along with other memory chip manufacturers, have been shifting production to high-bandwidth memory chips used in AI accelerators. This reduced the supply of standard DRAM chips, which are still needed for servers. “…in the medium term, we’re only able to meet about 50% to two-thirds of our demand from various key customers,” Micron CEO Sanjay Mehrotra said during the most recent earnings call.
Micron is ramping up its capital spending in 2026, but building memory chip facilities could take years. The first of two new factories in Idaho is scheduled to begin producing wafers around mid-2027, with the second expected to be operational by 2028. Meanwhile, a planned factory in New York is not expected to enter production until 2030.
Simply put, significant growth in supply is unlikely to occur in 2026. Micron will sell every memory chip it can produce, and prices are likely to continue to rise until either supply catches up or demand moderates. For now, Micron finds itself in a best-case scenario.
Micron’s revenue increased 57% year over year in the first quarter of fiscal 2026, and net income nearly tripled. While Micron’s production capacity is a major limitation, strong pricing driven by increasing demand for AI accelerators and servers will likely drive another year of strong revenue and earnings growth in 2026.
As the AI boom began, data center spending moved from standard components, such as CPUs, to more expensive AI accelerators. Old CPUs were not being upgraded as often as capital was channeled into expanding AI computing capacity.
This situation created a huge headwind for Intel, in addition to the loss of market share for AMD. However, the demand picture for server CPUs appears to be improving. At the Barclays Global Technology Conference in December, Intel noted that demand for server CPUs was surprisingly strong. Intel has already been shifting manufacturing capacity from PC CPUs to server CPUs, but the company still expects not to have enough supply to meet demand in early 2026.
There are a few reasons why the demand for server CPUs is increasing. First, hyperscalers are catching on to some extent, replacing servers powered by old, inefficient CPUs with newer models. AI servers are energy intensive, so more efficient components can significantly reduce the total cost of ownership.
Second, CPUs have a role to play in certain AI workloads. Retrieval-augmented generation, which combines large language models with external data sources to improve results without additional refinement, is one example. Intel’s latest server CPUs can handle aspects of the RAG pipeline thanks to their built-in AI processing power.
The latest generation of Intel server CPUs is manufactured on the Intel 3 process, and its next generation chips, scheduled to be launched in 2026, will use the Intel 18A process. As these processes increase, Intel will have more manufacturing capacity to meet the increasing demand for server CPUs.
Intel’s data center and AI segment saw a slight drop in revenue in the third quarter, but there could be a rebound on the horizon as Intel prioritizes server CPU production. While it is difficult to predict how long the server boom will last, speaking of the production of the AI bubble, Micron and Intel will benefit in 2026 from strong demand.
Before you buy stock in Micron Technology, consider this:
the Motley Fool Stock Advisor a team of analysts has identified only what they believe they are 10 best stocks for investors to buy now… and Micron Technology was not one of them. The 10 stocks that made the cut could produce monster returns in the coming years.
Consider when Netflix I made this list on December 17, 2004… if you invested $1,000 at the time of our recommendation, you would have $493,290!* Or when Nvidia I made this list on April 15, 2005… if you invested $1,000 at the time of our recommendation, you would have $1,153,214!*
Now, it is worth noting Stock consultant total average income is 973% — a market-crushing outperformance compared to 195% for the S&P 500. Don’t miss the latest top 10 list, available with Stock Advisorand join an investment community built by individual investors for individual investors.
See the 10 stocks »
*The Stock Advisor returns from January 6, 2026.
Timothy Green holds positions in Intel. The Motley Fool has positions in and recommends Advanced Micro Devices and Intel. The Motley Fool has a disclosure policy.
2 AI Stocks Riding A Server Boom That’s Accelerating Faster Than Anyone Expected was originally published by The Motley Fool
