The NAND Boom in the Era of AI Inference: Western Digital (SNDK.US) Investor Day May Reveal a New Growth Curve! The Storage Bull Market Awaits the Next Fundamental Catalyst
Any signals of growth prospects or indications of NAND demand and technological iteration released by management during the "SanDisk Investor Day" could serve as a fundamental confirmation for the continued rally in global storage stocks.
According to Zhitong Finance APP, the SanDisk Investor Day on August 13 may become a significant catalyst for the storage chip sector and even the global AI computing power industry chain. The US NAND storage chip giant SanDisk (SNDK.US) previously reported second quarter (company’s Q4) results that exceeded expectations across the board, but its revenue and profit outlook for the next quarter was slightly lower than the market's increasingly upgraded expectations. This highlights the continued expansion in demand for enterprise-grade NAND storage chips used in AI data centers, but market expectations for SanDisk's future growth are becoming increasingly stringent—any small shortfall could potentially trigger sharp stock price fluctuations. Any signals regarding strengthened growth prospects or NAND demand and technological iteration released by management at “SanDisk Investor Day” could serve as a “fundamental confirmation trigger” for the ongoing rebound in global storage stocks.
At last week's earnings conference call, SanDisk management’s outlook was not significantly above expectations. Management guided that gross margin would reach a plateau, maintaining around 83%–85%, corresponding to full-year EPS of around $180. Based on a 10x price-to-earnings ratio, the corresponding share price would be about $1,800. Currently, SanDisk's stock price hovers around $1,350. Essentially, the market does not believe that a gross margin of over 80% is sustainable; instead, it is pricing in a lower future EPS.
For at least the next two quarters, exceeding core performance expectations has been priced in by the market. Now, the focus has shifted: once NAND flash prices stop rising and supply gradually catches up with demand, can management reveal what the normalized gross margin level will be and deliver a strong signal regarding growth prospects for 2027, or possibly disclose technical iteration signals for one of the next big waves of structural demand for NAND—HBF.
More specifically, investors will focus on this signal: whether the company's new business model contracts based on LTA (i.e., long-term NAND supply agreements) can continue to support high gross margins and the current ever-strengthening enterprise-level data center NAND business revenue structure. Only then will investors believe that the stock warrants a substantial revaluation. Before such a signal emerges, any weekly rally may just be volatility disguised as an uptrend. If the fundamentals do not change, some analysts do not believe the stock price can return to historic highs.
Kioxia, SanDisk, SK hynix, and Samsung have explicitly defined High Bandwidth Flash (HBF) as a new NAND form factor designed to address the AI “memory wall,” aiming to provide greater capacity in AI inference, and claim that HBF can achieve near “infinite capacity HBM” aggregate transfer performance in related inference tests, while significantly increasing available memory capacity. SanDisk and SK hynix have published the first OCP technical specifications, attempting to place high-capacity, persistent NAND closer to AI accelerators to alleviate the “memory wall” during the inference phase, thereby increasing bandwidth and lowering overall token service costs while offering greater capacity than HBM; it is not a short-term replacement for HBM, but may instead create a new near-compute storage tier between HBM and traditional SSDs.
Key Validation of the “Inference Era” for NAND! Will SanDisk Investor Day Directly Address HBF Commercialization?
SanDisk (i.e., “SanDisk”) Investor Day could indeed become a pivotal catalyst for global storage chip trading and the AI computing infrastructure theme, but what the market really wants to hear is not “how much higher prices can go next quarter” but three things—normalized NAND gross margins, profitability stability of long-term supply agreements (LTA/NBM), and the commercialization timetable for HBF (High Bandwidth Flash) moving from technical standard to mass adoption.
Earlier on August 3, SanDisk and SK hynix jointly released the first OCP HBF technical specification, with AI leaders like Google and Tenstorrent participating in verification, positioning HBF as a new memory tier in AI inference systems close to the xPU, balancing high capacity and high bandwidth; on August 12, they also launched a new generation 2Tb QLC 3D NAND with Kioxia for AI infrastructure.
HBF (High Bandwidth Flash) is likely to become one of the largest structural drivers of demand for NAND in the next cycle—some analysts even stress that the latest HBF roadmap and outlook signal a “NAND attempt to replicate the HBM historic moment.” SanDisk and SK hynix released the first open HBF technical specification this August: it is based on 3D NAND but is no longer just for remote SSD storage; instead, it is placed closer to CPU/GPU/xPU using advanced packaging and UCIe interface, with a single package supporting up to 512GB capacity and 0.4—3.0TB/s bandwidth. SanDisk plans to offer initial HBF samples in the second half of 2026, with the first AI inference devices equipped with HBF entering the sample stage in early 2027.
Therefore, it is highly likely that management will further explain the HBF roadmap, mass production/customer introduction timeline, eSSD and QLC expansion, as well as how the new business model will sustain high gross margins at Investor Day. However, investors should be cautious: before official announcements at the event, “HBF technology roadmap and major commercialization announcements” should not be viewed as confirmed disclosures.
SanDisk’s own FYQ4 (i.e., fourth fiscal quarter) total revenue has reached $8.97 billion, a quarter-on-quarter increase of 51%, with around two-thirds of the growth attributable to pricing. It was also noted that total revenue from data center business rose approximately 400% year-on-year; if SanDisk management can further upgrade the “high ASP” into “long-term contracts + AI storage structural increments + HBF new product cycle,” it could further boost the strong rally in SanDisk and even global storage chip stocks.
More importantly, the current “AI storage bull market” already has solid cross-company and cross-medium industry chain evidence, not just SanDisk’s individual prosperity. Samsung clearly expects server DRAM, eSSD, and HBM demand to further accelerate in the second half of 2026, and even if smartphone/PC demand slows, the industry will remain under-supplied; SK hynix says Agentic AI is expanding the base demand for both DRAM and NAND storage chips, with both NAND and DRAM prices rising significantly in the second quarter, and they continue shifting NAND capacity toward 321-layer, high-capacity, and high-performance products.
Western Digital’s fourth-quarter revenue jumped 44% year-on-year to $3.75 billion, and Seagate’s total revenue increased 48% year-on-year to $3.63 billion, with free cash flow of $1.1 billion. Both companies emphasized the large capacity storage demand from AI cloud data centers; it is important to differentiate that Western Digital and Seagate are now mainly addressing HDD/mass-capacity storage needs, not directly proving NAND demand, but together they show that the AI inference era is creating a larger data lifecycle—from training data, model checkpoints, RAG databases, logs, inference context, to long-term data retention, all driving demand from HBM through DRAM, eSSD/NAND, and HDD. The South Korean stock market, where the two storage giants account for 50% weight, has risen more than 20% from its July 30 low, entering a technical bull market, underscoring that the HBM/DRAM/NAND storage super cycle driven by the AI infrastructure boom remains strong.
Morgan Stanley Bear Turns Bull, Validating That the Storage Super Cycle Has More Room to Run
For bullish sentiment on storage chip stocks, the shift of Morgan Stanley’s senior analyst Shawn Kim from a previous “bearish spokesperson” stance to viewing the correction as nearly over is a significant market signal. However, the more accurate interpretation is not “he suddenly thinks storage prices will always rise,” but rather that he feels the market has over-traded the “second derivative downturn in pricing.” Shawn Kim believes the sharpest short- and medium-term correction in memory chips is over, current valuations provide a tactical re-entry, and he has raised SK hynix’s 2026 EPS again.
Morgan Stanley’s research shows that Q3 NAND contract prices are expected to rise 20% quarter-on-quarter on a high base, with the industry continuing to shift capacity from consumer to eSSD; meanwhile, Kim now sees the next stage of share price drivers moving from simply ASP increases to “LTA long-term supply agreements + FCF growth trajectory + strong capital returns.” His latest views also reveal that the next phase of NAND investment logic is shifting from “NAND price beta” to “AI inference demand + eSSD content increase + long-term contract price locks + HBF new architecture alpha.”
From the perspective of AI system engineering, the biggest difference in this storage super cycle is that HBM/server DRAM/enterprise SSDs are moving from common cyclical products to the physical bottlenecks of AI computing power systems. Musk, in the SpaceX 2026 Q2 financial conference call, rarely commented on the storage market, saying that while storage supply grows about 20% annually, demand is growing at 200% or even higher, leading to a severe imbalance and making price increases an economic inevitability. SK hynix CEO Kwak Noh-Jung commented in a July conference call that 2027 may be the tightest supply year in global storage industry history, and client demand exceeding company supply capabilities may persist beyond 2030.
Disclaimer: The content of this article solely reflects the author's opinion and does not represent the platform in any capacity. This article is not intended to serve as a reference for making investment decisions.
You may also like
Forecasts for $1M Bitcoin price deemed too ambitious
London loses ground to Paris in battle for crypto conference crown
Chime Financial explores stablecoins for app integration as neobank joins Open Standard consortium
Three trusted npm libraries shipped poisoned versions that phone home through Ethereum
