Morgan Stanley has raised its price target for Silicon Motion to $400, pointing to a sharp tightening in the NAND memory market as artificial intelligence data centers consume rapidly growing amounts of storage. The bank expects global NAND supply to move from a small surplus in 2025 to deep shortages in 2026 and 2027, creating a stronger pricing environment for companies tied to enterprise SSD controllers, AI boot drives and data center storage infrastructure.
The call reflects a major shift in the memory market. Demand that was once led mainly by smartphones and personal computers is increasingly being driven by cloud infrastructure, enterprise solid-state drives and AI servers. Morgan Stanley estimates that NAND supply and demand will move from a 2% surplus in 2025 to a 15% shortage in 2026 and a 9% shortage in 2027, as data center purchasing outpaces industry output.
Total NAND demand is forecast to rise from 1,111 exabytes in 2025 to 1,484 exabytes by 2027. Supply, however, is expected to reach only 1,347 exabytes by that year. The gap suggests that the industry may struggle to meet the storage needs of AI workloads, even as memory producers lift output and prioritize higher-margin enterprise products.
The shortage is expected to have a direct effect on pricing. Morgan Stanley projects enterprise-level TLC SSD prices could rise by about 30% quarter over quarter by the third quarter of 2026. Consumer NAND products, by contrast, are expected to see weaker price increases because smartphone and PC makers face tighter margins and may resist passing higher component costs through to end customers.
AI demand changes the NAND market
The center of NAND demand is moving away from consumer electronics and toward AI infrastructure. According to Morgan Stanley’s forecast, AI-related NAND usage could reach 609 exabytes in 2027, accounting for 41% of total consumption.
That would mark a significant change in how NAND memory is used across the global technology supply chain. NAND is the nonvolatile memory used in SSDs, phones, PCs, servers and a wide range of connected devices. It stores data even when power is turned off, making it essential for both consumer devices and cloud systems.
In the AI era, however, the most important growth is coming from servers. Training and running large AI models require enormous volumes of data to be stored, retrieved and moved quickly. Data centers need high-performance SSDs for model training, inference, caching, database workloads and fast access to large datasets. They also need boot drives, which are smaller storage modules used to start and manage server systems inside cloud infrastructure.
This is reshaping market dynamics. Previously, smartphone shipments and PC replacement cycles played a larger role in determining NAND pricing. Now, spending by cloud service providers and enterprise customers is becoming a more powerful force. When major technology companies accelerate AI server construction, NAND demand can rise quickly and absorb supply that might otherwise flow into consumer markets.
Morgan Stanley’s analysis suggests this is not simply a short-term inventory cycle. The bank sees a structural shift in demand, with AI-related storage taking a larger share of output over the next several years. That shift may support higher prices for enterprise SSDs and related components, even if demand for phones and PCs remains uneven.
Silicon Motion benefits from enterprise SSD exposure
Silicon Motion is central to Morgan Stanley’s upgraded view because of its exposure to SSD controller chips and AI server storage products. SSD controllers are key components that manage how data is written to and read from NAND memory. As SSDs become faster and more complex, controller technology becomes more important, especially in enterprise and AI applications.
The company’s MonTitan enterprise SSD controller business is expected to become a larger contributor to revenue over the next several years. Morgan Stanley forecasts MonTitan will account for 5% of Silicon Motion’s total revenue in 2026, 13% in 2027 and 19% in 2028.
AI boot drive products are also expected to become a meaningful source of revenue. The bank estimates boot drive products could contribute 15% of total revenue in 2026 and 21% in 2027. That growth reflects increasing demand for storage modules used in cloud servers, where reliability, speed and energy efficiency are important purchasing factors.
The higher price target also reflects the broader pricing environment. If NAND shortages push enterprise SSD prices higher, controller suppliers with strong exposure to data center products may be able to benefit from both stronger demand and richer product mix. Traders have been watching this part of the semiconductor supply chain closely because it sits between NAND manufacturers and the cloud infrastructure market.
Bank of America has also shown optimism toward Silicon Motion, raising its price target to $450. It projected that the company’s revenue could exceed $1.5 billion in 2026, supported by demand for PCIe Gen5 controllers used in AI workloads. The firm’s 2026 earnings per share growth has been forecast at 135.8%, reflecting expectations for a powerful recovery in profitability if enterprise demand continues to rise.
Price increases are spreading unevenly
The NAND price recovery is not expected to affect all product categories equally. The strongest pricing appears likely in enterprise SSDs, where cloud companies and server customers have urgent capacity needs and greater ability to absorb higher costs.
Morgan Stanley expects enterprise TLC SSD prices to climb about 30% quarter over quarter by the third quarter of 2026. Other market projections cited in the analysis show an even sharper earlier move, with NAND flash contract prices rising 33% to 38% in the first quarter of 2026 and then projected to climb another 70% to 75% in the second quarter.
That type of pricing strength would be significant for memory producers and component suppliers. It would indicate that customers are competing for limited supply, especially in high-performance storage products used by data centers.
Consumer NAND products may not see the same gains. Smartphone and PC manufacturers operate in markets where end demand is more price sensitive. If NAND costs rise too quickly, device makers may reduce storage configurations, delay purchases or pressure suppliers for better terms. That could limit price increases in consumer channels even while enterprise products continue to climb.
This divergence is important because it shows that the NAND market is becoming more segmented. AI servers and enterprise SSDs are driving the tightest supply conditions, while consumer electronics may remain constrained by weaker demand and lower profitability. As a result, companies with stronger exposure to enterprise storage may outperform those more dependent on commodity consumer NAND products.
Longsys and Phison also receive higher targets
Morgan Stanley also raised price targets for Longsys and Phison, two Asian companies exposed to the storage supply chain. Longsys’ price target was lifted to 673 yuan, while Phison’s was raised to 2,588 New Taiwan dollars. Both companies kept an “Equal Weight” rating.
The higher targets reflect improved pricing conditions and stronger demand across NAND-related products. Phison, a major NAND controller and storage solutions provider, has seen revenue forecasts for fiscal 2026 increase sharply. Analysts’ consensus price target has also risen, with revenue forecasts for that year reportedly lifted by 59%.
Longsys, meanwhile, has already seen a major stock market reaction to the tightening storage cycle. Its share price has risen 768.90% over the past year, reflecting strong enthusiasm around NAND pricing, enterprise storage and demand from AI-related applications.
Even so, Morgan Stanley appears more cautious on Longsys and Phison than on Silicon Motion. One reason is supply access. During periods of tight NAND availability, companies that do not control core NAND production may face allocation limits. They may benefit from rising prices, but they may also struggle to secure enough supply to maximize revenue growth.
This means pricing power does not always translate directly into higher sales volumes. For module makers and controller suppliers, access to NAND wafers and close relationships with major memory producers can be just as important as end demand. If supply is rationed, some companies may have less flexibility than large integrated memory manufacturers.
Long-term contracts become more important
As shortages form, long-term procurement agreements are becoming a key tool for managing price risk and supply access. Large customers want to secure memory supply before shortages worsen, while suppliers want visibility into future demand and pricing.
Morgan Stanley estimates that Kioxia’s coverage of long-term procurement contracts could exceed 50% by 2027. That would give the company greater revenue visibility and may help stabilize pricing during a volatile period.
Micron’s contracts are also important because they include price floors and ceilings. These terms can reduce volatility by preventing prices from falling too far in weak markets or rising too far during periods of intense demand. For customers, this can protect budgets. For suppliers, it can support planning and reduce the danger of sudden pricing collapses.
However, these contracts can also limit upside during a shortage. If spot prices or new contract prices surge, suppliers locked into price ceilings may not fully capture the highest market prices. At the same time, customers with secured supply may gain an advantage over competitors forced to buy in a tighter open market.
The spread of long-term agreements shows how strategic NAND has become. For AI data centers, storage is no longer a routine component purchase. It is part of the infrastructure required to deploy and scale AI services. Cloud providers that cannot secure enough storage may face delays in server deployment or higher operating costs.
Cloud spending is the main swing factor
The outlook for NAND now depends heavily on capital expenditure by major cloud service providers. These companies are building the AI infrastructure that consumes large volumes of high-performance storage. Their spending plans influence demand for servers, GPUs, networking equipment, SSDs and power systems.
If cloud companies continue to raise AI infrastructure budgets, NAND demand could remain stronger for longer. That would support the shortage case and help sustain higher prices for enterprise SSDs and related components.
If spending slows, the shortage may ease faster. AI infrastructure projects are expensive, and cloud providers must balance growth ambitions with profitability, power availability and customer demand. Any moderation in AI server buildouts could reduce pressure on NAND supply.
This makes cloud capital expenditure one of the most important indicators for traders following memory stocks. Announcements from major cloud companies can quickly change expectations for storage demand. Stronger data center spending may support earnings forecasts for memory and controller companies, while slower spending may raise concerns about oversupply.
The NAND market is also tied to the broader semiconductor cycle. Memory products have historically moved through boom-and-bust periods. Prices rise when supply is tight, leading producers to expand capacity. Those expansions can later create too much supply, causing prices and margins to fall. AI demand may extend the current upcycle, but it does not remove the risk of a future reversal.
Decentralized storage could feel the impact
Rising demand for data hosting is also supporting interest in decentralized storage networks. This sector is expected to grow from about $856 million in 2026 to more than $5 billion by 2035, according to projections cited in the market discussion.
Decentralized storage networks use distributed hardware rather than relying only on traditional centralized data centers. They can support applications in Web3, content delivery, archival storage and privacy-focused data services. While their models differ from large cloud providers, they still depend on physical storage hardware.
That means NAND pricing matters for this emerging sector as well. If SSD prices rise sharply, the cost of expanding decentralized storage capacity can increase. Higher hardware costs may affect network economics, node profitability and the pace of deployment.
At the same time, greater demand for alternative storage models may reinforce the broader storage growth story. As AI, cloud computing and decentralized platforms all require more data capacity, NAND becomes a critical layer of digital infrastructure.
Risks could rise after 2028
Morgan Stanley’s shortage scenario may not last indefinitely. The bank’s testing scenarios suggest NAND conditions could reverse after 2028, especially if new capacity comes online and AI-related capital expenditure slows.
One major variable is YMTC, China’s leading NAND producer. If YMTC raises capacity to 470,000 wafers per month, the market could move from shortage to balance, or even surplus. That would change the pricing outlook significantly.
A supply recovery would be especially important if smartphone and PC makers continue resisting higher component costs. Consumer device makers may not accept sustained NAND price increases if end-market demand remains soft. If enterprise demand slows at the same time that new capacity arrives, the market could swing quickly from tightness to oversupply.
This is the classic risk in memory markets. Tight supply encourages expansion. Expansion eventually increases output. If demand growth does not keep pace, the cycle turns. Prices then fall, margins compress and earnings expectations are revised lower.
For now, Morgan Stanley’s view is that AI storage demand is strong enough to reshape the NAND market in 2026 and 2027. The projected shortages are large, and the demand mix is shifting toward higher-value enterprise products. That creates a favorable setup for selected companies in the storage supply chain, especially those tied to enterprise controllers and AI server modules.
But the longer-term balance remains uncertain. It will depend on how quickly YMTC expands, how disciplined other NAND producers remain, how much cloud companies spend on AI data centers and whether consumer electronics demand can support higher costs.
For traders, the NAND market is becoming a more important signal for the broader AI hardware trade. GPUs and networking chips have received much of the attention, but storage is now emerging as another critical bottleneck. If AI systems cannot store and retrieve data efficiently, the rest of the infrastructure cannot operate at full potential.
Morgan Stanley’s $400 target for Silicon Motion reflects that changing reality. The bank’s analysis suggests that the next phase of AI infrastructure growth may reward companies that provide the less visible but essential hardware behind data movement and storage. In a market where AI demand is absorbing supply faster than producers can add capacity, NAND controllers, enterprise SSDs and boot drive modules are moving closer to the center of the semiconductor growth story.
As AI reshapes storage demand, sharpen your crypto strategy with AI copy trading insights tailored for data-driven markets.
Disclaimer: The content on this page is provided for general informational purposes only and does not represent the views or financial advice of Toobit. We make no guarantees regarding the accuracy or completeness of this information and shall not be held liable for any errors, omissions, or outcomes resulting from its use. Investing in digital assets involves risk; users should independently evaluate their financial situation and the risks involved. For further details, please consult our Terms of Service and Risk Disclosure.
