Demand for memory for AI systems is driving exceptional growth figures at Micron. The memory chip manufacturer saw its quarterly revenue nearly quintuple in one year to $54.2 billion. The company also expects further growth in the new fiscal year.
Micron closed its fiscal fourth quarter with revenue of $54.23 billion. A year earlier, that figure was $11.32 billion. Net income rose from $3.2 billion to $37.7 billion during the same period. For the full fiscal year, revenue totaled $133.19 billion, compared to $37.38 billion a year earlier.
This strong growth is primarily linked to the rapid expansion of AI infrastructure. Data centers require large amounts of high-speed memory to supply data to GPUs. As a result, high-bandwidth memory (HBM) in particular has become scarce. Micron is benefiting not only from higher volumes but also from higher memory prices.
This is reflected in the margins. Micron reported a gross margin of nearly 87 percent for the past quarter. The Core Data Center division even achieved a gross margin of 90 percent and saw its revenue rise to $18 billion. The Cloud Memory division accounted for another $16.3 billion.
Memory is becoming more important for AI
CEO Sanjay Mehrotra expects the role of memory to continue growing as AI systems become larger and more complex. Micron is therefore investing further in new memory technology and additional production capacity.
A key part of this is HBM. This memory is placed next to GPUs and offers much more bandwidth than standard DRAM. As a result, it has become a crucial component of AI accelerators. In this market, Micron competes primarily with South Korea’s SK hynix and Samsung.
According to SiliconANGLE, Micron is also working with Nvidia on a customized HBM implementation. The collaboration is part of a broader trend in which memory manufacturers are tailoring their products more and more specifically to future AI accelerators. Micron is now producing HBM4 for Nvidia’s Vera Rubin platform and has previously indicated its intention to supply custom HBM for the Feynman generation starting in 2028.
Shortages persist
Meanwhile, the rapid growth of AI is having consequences beyond the data center. Manufacturers are struggling to keep up with the demand for DRAM and other types of memory. As a result, prices have risen sharply, and manufacturers of PCs, smartphones, and other electronics are also having to pay more for memory.
Micron is working to expand its production capacity. Among other things, the company is investing in new factories in Idaho and New York. At the same time, the outlook remains exceptionally strong. For the first quarter of fiscal year 2027, Micron expects revenue of approximately $61.5 billion and adjusted earnings of $38.15 per share. The expected gross margin also remains at a very high level, at over 86 percent.
As a result, there are currently few signs of a market slowdown. While the memory sector has historically been a highly cyclical part of the chip market, the development of AI infrastructure is currently driving a combination of supply shortages, high prices, and long-term agreements with major customers.