Micron Technology forecast first-quarter revenue well above Wall Street expectations on Wednesday as demand for high-bandwidth memory used in artificial-intelligence data centers continued to outstrip supply. The company said customers had increased financial commitments under long-term supply agreements to $32 billion, up from $22 billion in June, providing one of the clearest signs yet that the AI infrastructure boom is reshaping the memory-chip market.
Micron expects revenue of about $61.5 billion, plus or minus $1.5 billion, compared with an analyst consensus of $57.02 billion. The company also said remaining performance obligations under long-term agreements had risen to roughly $150 billion, up from about $100 billion in the prior quarter.
Micron AI Memory Demand Is Outrunning Capacity
The Micron AI memory boom is being driven by high-bandwidth memory, or HBM, a specialized form of DRAM that allows AI processors to move large amounts of data quickly. Nvidia and other chipmakers rely on HBM to keep advanced accelerators supplied with data during model training and inference.
Micron President and Chief Operating Officer Manish Bhatia said memory has become the chief constraint in AI infrastructure, ahead of several other bottlenecks that have received more attention. The company has seen orders exceed available capacity and expects supply-demand conditions to become even tighter in fiscal 2027 and 2028.
That scarcity is changing how large customers buy chips. Instead of relying mainly on short-term orders, cloud companies and other buyers are committing cash deposits and multiyear agreements to secure future supply.
Customers Have Committed $32 Billion to Supply Deals
CEO Sanjay Mehrotra said the company’s long-term customer commitments have risen by $10 billion since June. Most of those agreements include cash deposits, giving Micron more visibility into future demand and reducing some of the volatility that historically defined the memory industry.
The company has already secured agreements for most of its 2027 HBM output. That means customers are reserving advanced memory well before new factories and production lines are fully available, a sign that buyers expect AI data-center spending to remain elevated.
The shift is important because memory chips were traditionally among the most cyclical semiconductor products. Prices could swing sharply when manufacturers added too much capacity, but long-term contracts can make revenue more predictable and reduce the risk of sudden oversupply.
HBM Has Become a Critical AI Bottleneck
HBM is physically stacked and connected in ways that allow much faster data transfer than conventional memory. That design is particularly important for AI accelerators, where the processor can become underutilized if memory cannot deliver data quickly enough.
American News Brief previously reported that Samsung expects the AI chip shortage to remain tight through 2028, with HBM capacity consuming an increasing share of global DRAM production. Micron’s new forecast reinforces that broader shortage story from the perspective of the only major U.S.-based HBM producer.
The company competes with South Korea’s SK Hynix and Samsung Electronics, which currently lead global HBM market share. All three are racing to expand capacity, but new production takes years to build and several quarters to ramp efficiently.
Micron Is Expanding U.S. Manufacturing
Micron plans to invest more than $250 billion in the United States through 2035, including manufacturing expansions in New York, Idaho and Virginia. The company raised that commitment in July as AI demand accelerated and Washington continued to prioritize domestic semiconductor production.
The U.S. investment plan is strategically significant because advanced memory production is concentrated in a small number of countries and companies. Expanding domestic capacity could reduce supply-chain dependence while giving U.S. technology firms more secure access to components needed for AI systems.
Micron has said its first new silicon wafer output from current expansion projects is expected in mid-2027. Even then, executives cautioned that meaningful market impact will take additional quarters as factories ramp toward full utilization.
Fourth-Quarter Results Also Beat Expectations
The strong guidance followed a fourth quarter in which revenue more than quadrupled to $54.23 billion, topping analysts’ estimate of $51.07 billion. Adjusted profit reached $33.42 per share, also above expectations.
For the first quarter, Micron expects adjusted profit of about $38.15 per share, plus or minus $1, compared with the $35.40 consensus. The size of the forecast beat suggests that pricing power and product mix remain favorable even after the company’s stock more than tripled this year.
Micron joined the $1 trillion market-value club in May, reflecting investor expectations that AI demand could produce a longer and more profitable memory cycle than previous technology booms. The company’s latest results provide evidence supporting that thesis, but they also raise expectations for continued execution.
AI Spending Remains the Key Risk and Opportunity
The biggest opportunity is continued capital spending by cloud and technology companies. Amazon, Alphabet and other major firms are expected to spend hundreds of billions of dollars this year on AI infrastructure, creating enormous demand for processors, networking equipment and memory.
The main risk is that customers eventually slow those investments or find ways to use memory more efficiently. A sharp change in AI spending could leave manufacturers with expensive new capacity and revive the oversupply cycles that have historically hurt the industry.
For now, customer commitments suggest buyers are more worried about shortages than excess inventory. Micron’s backlog, deposits and 2027 HBM reservations indicate that advanced memory is becoming one of the most valuable and strategically important components in the global AI buildout.
