Micron investors take fresh comfort as Wall Street lifts forecasts and Nvidia’s Jensen Huang warns memory supply may stay tight for years

Micron Technology is drawing renewed investor attention as Wall Street sentiment improved after the company’s latest results and as Nvidia CEO Jensen Huang renewed the case for a prolonged memory-chip shortage. In commentary circulating in markets coverage, analysts said forward expectations are rising, while Huang pointed to continuing, outsized demand tied to AI infrastructure builds.

The optimistic framing centers on the persistence of demand for both DRAM (dynamic random-access memory, used as fast working memory for computing) and NAND flash (used for storage). Memory chips are a critical input into AI systems, and the market has been closely watching whether supply expansion can keep pace with the scale-up of data-center spending.

Micron-related coverage highlighted that the company recently announced 16 multi-year customer deals. Multi-year contracts are relatively uncommon in the memory industry, where pricing and supply conditions can swing. The markets narrative described these agreements as evidence that customers are willing to lock in capacity over longer periods, potentially smoothing Micron’s near-term outlook.

The same coverage also said several analysts increased their forward earnings estimates after Micron’s most recent report. One cited industry view expected earnings growth to reach 168% annually through fiscal 2027, and it framed the stock’s recent surge as a response to that combination of higher forecasts and a tighter supply-demand balance.

Nvidia’s role in the story is the demand announcement. Huang is described as telling investors that the memory shortage could continue for several years, attributing the imbalance to the magnitude of AI demand for systems that need large amounts of memory bandwidth and capacity. Nvidia’s comment matters in this context because Nvidia’s accelerators are central to training and running AI workloads, and memory needs scale with deployments.

Micron’s business exposure is also tied to a specific memory category used in AI servers, high-bandwidth memory (HBM). HBM is a specialized form of DRAM designed to deliver higher data throughput to processors such as GPUs. Commentary noted that both HBM and other memory types are integral for AI workloads, which helps explain why memory suppliers have become a focal point as data-center capex has accelerated.

Still, key details remain unconfirmed in the available market commentary. The coverage did not provide the full terms of Micron’s 16 multi-year deals, including durations, pricing mechanics, or how much of Micron’s total output each agreement represents. It also did not show the specific figures behind the analyst forecast revisions, such as whether changes were driven by gross margin expectations, volume assumptions, or both. As with much of the AI trade, the direction of travel is clearer than the exact numbers.