Core Viewpoint - The rapid development of AI has made storage a critical component in the AI infrastructure, alongside computing power. The demand for storage is surging due to the increasing data volume and inference scenarios driven by large models and generative AI. Three storage technologies—HBM, HBF, and GDDR7—are redefining the future landscape of AI infrastructure [1]. Group 1: HBM (High Bandwidth Memory) - HBM has evolved from a high-performance AI chip component to a strategic point in the storage industry, significantly impacting AI chip performance limits. In less than three years, HBM has achieved over twofold capacity and approximately 2.5 times bandwidth increase [3]. - SK Hynix is leading the HBM market, currently in the final testing phase for the sixth generation (HBM4) and has announced readiness for mass production. In contrast, Samsung is facing challenges in HBM4 supply to Nvidia, with a two-month delay in testing [3][5]. - A notable trend is the customization of HBM, driven by cloud giants developing their AI chips. SK Hynix is shifting towards a fully customized HBM approach, collaborating closely with major clients [4]. Group 2: HBF (High Bandwidth Flash) - HBF aims to address the limitations of traditional storage by combining the capacity of NAND flash with the bandwidth of HBM. Sandisk is leading the development of HBF technology, which is expected to meet the growing storage demands of AI applications [8][9]. - HBF is seen as complementary to HBM, suitable for specific applications requiring large block storage units. It is particularly advantageous in scenarios demanding high capacity but with relatively relaxed bandwidth requirements [10][11]. Group 3: GDDR7 - Nvidia's introduction of the Rubin CPX GPU, utilizing GDDR7 instead of HBM4, reflects a new approach to AI inference architecture. This design optimizes resource allocation by separating the inference process into two stages, effectively utilizing GDDR7 for context building [13]. - The demand for GDDR7 is increasing, with Samsung successfully meeting Nvidia's orders. This flexibility positions Samsung favorably in the graphics DRAM market [14]. - GDDR7's cost-effectiveness may drive the widespread adoption of AI inference infrastructure, potentially increasing overall market demand for high-end HBM due to the proliferation of applications [15]. Group 4: Industry Trends and Future Outlook - The collaborative evolution of storage technologies is crucial for the AI industry's growth. HBM remains essential for high-end training and inference, while HBF and GDDR7 cater to diverse market needs [23]. - The ongoing innovation in storage technology will accelerate as AI applications expand across various sectors, providing tailored solutions for both performance-driven and cost-sensitive users [23].

Core Viewpoint - The article discusses the explosive growth potential in the memory market, driven by AI and data center demands, highlighting a "super cycle" in memory pricing and production, particularly for DRAM, HBM, and NAND [2][11][23]. Market Trends - The storage market is experiencing upward trends, with significant price increases in DDR and NAND due to supply chain disruptions and rising demand from AI applications [2][8]. - Recent reports indicate that Micron has raised its server shipment growth forecast for 2025 to approximately 10%, driven by increased demand for AI agents and traditional server workloads [9]. - TrendForce predicts a 5-10% average price increase for NAND Flash products in Q4 due to supply shortages and rising demand from cloud service providers [10]. Price and Profitability Drivers - Key drivers of the current memory super cycle include: 1. Explosive demand for AI and data centers, with traditional server capital expenditures expected to grow by 20-30% by 2026, leading to a 50% increase in DDR4/DDR5 memory demand [14]. 2. Profit margins for DRAM are projected to rise from 40-50% to nearly 70% by 2026, while NAND margins are expected to improve from breakeven to 30-40% [14]. Demand Surge Factors - The recent surge in storage demand is attributed to the transition of AI applications from an "accumulation phase" to a "high penetration phase," significantly increasing user interaction and data generation [19]. - The upgrade in AI technology logic has also amplified the need for DRAM and NAND, with token consumption increasing dramatically due to more complex interactions and multi-modal data processing [20]. - Companies are restructuring their AI infrastructure to implement a tiered storage system, which is driving immediate demand for DRAM and NAND products [21]. Future Outlook - The AI-driven super cycle is expected to last at least until 2027, with potential downturns anticipated in 2028 [23]. - Ongoing negotiations between DRAM manufacturers and NVIDIA regarding HBM pricing are likely to favor DRAM manufacturers, potentially leading to higher growth predictions for the HBM market [25]. Technological Developments - NVIDIA's introduction of the CPX solution is expected to create differentiated demand across storage products, potentially increasing GDDR7 demand while impacting HBM4 negatively in the short term [27]. - NVIDIA is also developing HBF (High Bandwidth Flash) as a cost-effective alternative to HBM, indicating a strategic shift in memory resource allocation [28].

AI 的火热不只是让人们争相讨论大模型和算力芯片，也彻底点燃了对内存的需求。过去几年，HBM（高带宽内存）成为了这场浪潮里最受追捧的「隐形明 星」。没有它，就没有英伟达 A100、H200 以及其他 AI 芯片的爆火，也也不会有无数大模型在短时间内跑出来并且迅速迭代。 正因如此，HBM 供不应求，几乎成了半导体行业的「硬通货」，也让 HBM 主要厂商 SK 海力士一举超越三星，成为全球最大存储芯片制造商。 | Ranking | Company | | Revenue (USSM) | | Market Share | | | --- | --- | --- | --- | --- | --- | --- | | | | 2Q25 | 1Q25 | QoQ | 2Q25 | 1Q25 | | 1 | SK hynix | 12,229 | 9,718 | 25.8% | 38.7% | 36.0% | | 2 | Samsung | 10,350 | 9,100 | 13.7% | 32.7% | 33.7% | | 3 | Micron | 6,950 | 6,575 | 5.7% | 22.0% | 24 ...

Summary of the Conference Call on NAND Industry and AI Impact Industry Overview - The conference call focuses on the NAND industry and its intersection with artificial intelligence (AI) technology, highlighting the expected growth and changes in market dynamics due to AI applications [1][2][9]. Key Insights - **AI's Entry into NAND**: The demand for NAND is expected to surge due to AI inference, with the overall addressable market projected to reach $29 billion by 2029. AI NAND is anticipated to account for 34% of the global NAND market by 2029 [1][2][9]. - **Supply Shortages**: A supply shortage in the NAND industry is expected to begin in the second half of 2026, following a prolonged period of oversupply since 2022. This shift is attributed to increased demand from cloud service providers and limitations in HDD capacity [2][10]. - **Market Dynamics**: The transition from traditional NAND to newer technologies like QLC eSSD and NL SSD is expected to reshape the market. The latter is projected to capture 5% of the market share by 2026, potentially leading to an 8% NAND shortage [2][10]. Company-Specific Insights - **SanDisk Corporation (SNDK.O)**: Target price raised from $70 to $96, identified as a top pick due to its favorable positioning in the AI NAND market [1][3]. - **KIOXIA and Samsung Electronics**: Both companies are expected to benefit significantly from the long-term growth trends in the NAND market [3][17]. - **Component Suppliers**: Companies like Phison and Longsys are rated overweight due to anticipated improvements in NAND pricing as supply tightens [3][17]. Additional Considerations - **Technological Innovations**: The introduction of high-bandwidth flash (HBF) and NL SSD technologies is expected to alleviate memory capacity bottlenecks in AI clusters, further driving NAND demand [2][12]. - **Earnings Projections**: The average earnings forecast for the preferred stocks in the NAND sector is projected to exceed market consensus by 26% for 2026, indicating a strong outlook for AI NAND growth [3][14]. - **HDD Market Impact**: The rise in AI-driven data requirements is expected to benefit HDD manufacturers like WDC and STX, as the total cost of ownership between HDD and SSD remains significantly different [3][17]. Conclusion - The NAND industry is poised for a significant transformation driven by AI applications, with expected supply shortages and technological advancements creating new investment opportunities. Companies like SanDisk, KIOXIA, and Samsung are well-positioned to capitalize on these trends, while component suppliers are also likely to see favorable market conditions.

Summary of the Conference Call on NAND Industry and AI Impact Industry Overview - The NAND industry is experiencing a resurgence due to the increasing demand from AI inference applications, with an estimated incremental Total Addressable Market (TAM) of **US$29 billion** by **2029** [1][13][22]. - The AI NAND market is projected to account for **34%** of the global NAND market by **2029**, with AI-related NAND usage expected to reach **431EB**, representing **20%** of global NAND bit shipments [13][22]. Key Insights - The demand for high-speed, high-capacity storage solutions, particularly QLC eSSD, is becoming critical for AI applications, as it meets the requirements for randomized I/O access [12][21]. - A NAND shortage is anticipated to emerge in the second half of **2026**, driven by a combination of AI infrastructure deployment and consumer demand weakness [14][24]. - Major Cloud Service Providers (CSPs) are already in discussions with suppliers for large orders of AI NAND and nearline SSDs, indicating a shift in procurement strategies [14][24]. Market Dynamics - The NAND industry has been in oversupply since **2022**, but a more balanced supply-demand dynamic is expected moving into **2026** [14][24]. - The introduction of nearline SSDs is expected to take market share from HDDs, with a potential **5%** market share shift translating into an **8%** NAND shortage in a bullish scenario [13][23]. Stock Recommendations - **SanDisk** has been elevated to a Top Pick with a price target raised from **$70** to **$96**, reflecting confidence in its ability to capitalize on the AI-driven demand [12][41]. - Other recommended stocks include **KIOXIA** and **Samsung Electronics**, which are expected to benefit significantly from the growth in the NAND sector [15][27]. - Within Greater China, module makers **Phison** and **Longsys** are favored due to better NAND pricing tailwinds [15][27]. Financial Projections - The 2026 EPS estimates for key recommended stocks are projected to be **26%** higher than consensus estimates, indicating a bullish outlook on the AI NAND growth story [30][41]. - SanDisk's EPS is expected to reach **$8.64** next year, compared to consensus estimates of **$6.11**, suggesting that the stock is undervalued given the potential for improved earnings [41][48]. Risks and Considerations - There is caution regarding potential inventory corrections in early **2026**, as demand may not fully overshadow the PC/mobile markets in the near term [47]. - The NAND industry is facing challenges from conservative capital expenditure plans and geopolitical tensions, particularly concerning China [35][44]. Conclusion - The NAND industry is poised for significant growth driven by AI applications, with key players like SanDisk, KIOXIA, and Samsung Electronics positioned to benefit. The anticipated shortage in **2026** and the shift towards eSSD solutions present compelling investment opportunities, although caution is warranted regarding potential market corrections and geopolitical risks.

Core Insights - High Bandwidth Memory (HBM) is emerging as a critical component for AI model training and inference due to its unique 3D stacked structure, which significantly enhances data transfer rates compared to traditional memory solutions like GDDR [1][2] HBM Market Dynamics - SK Hynix has established a dominant position in the HBM market, with its market share surpassing that of Samsung, which has seen a decline from 41% to 17% in the same period [3][4] - The launch of HBM3E has been pivotal for SK Hynix, attracting major tech companies like AMD, NVIDIA, Microsoft, and Amazon, leading to a significant increase in demand [3] - SK Hynix's sales in DRAM and NAND reached approximately 21.8 trillion KRW, surpassing Samsung's 21.2 trillion KRW for the first time [3] Competitive Landscape - Samsung is attempting to regain its footing by reviving Z-NAND technology, aiming for performance improvements of up to 15 times over traditional NAND and a reduction in power consumption by up to 80% [6][7] - NEO Semiconductor has introduced X-HBM architecture, which offers 16 times the bandwidth and 10 times the density of existing memory technologies, targeting the AI chip market [10] - Saimemory, a collaboration between SoftBank, Intel, and Tokyo University, is developing a new stacked DRAM architecture aimed at becoming a direct HBM alternative with significant performance improvements [11] Innovations and Alternatives - SanDisk and SK Hynix are collaborating on High Bandwidth Flash (HBF), a new storage architecture designed for AI applications, which combines 3D NAND flash with HBM characteristics [12][13] - The industry is exploring various architectural innovations, such as Processing-In-Memory (PIM), to reduce reliance on HBM and enhance efficiency [15][16] Future Trends - The AI memory market is expected to evolve into a heterogeneous multi-tiered structure, where HBM will focus on training scenarios, while PIM memory will cater to high-efficiency inference applications [18] - The demand for HBM, particularly HBM3 and above, is projected to remain strong, with significant price increases noted in the market [17]

Core Viewpoint - NAND manufacturers are often overlooked in the AI boom, as their growth potential in AI applications is limited compared to DRAM, particularly HBM, which is essential for AI computing needs [1][2][3] NAND Market Dynamics - The NAND market has experienced significant fluctuations, with a notable recovery in early 2024 driven by increased demand from the smartphone sector and AI server construction [6][8] - Global smartphone shipments rose by 8.2% year-on-year in Q1 2024, with a shift towards larger capacity models [6] - AI server demand led to a 32% quarter-on-quarter increase in enterprise SSD procurement, with North America's top three cloud service providers accounting for 58% of global orders [6][7] - Strategic production cuts by major manufacturers have resulted in a 17% increase in NAND Flash contract prices in Q1 2024 [6][9] AI Demand and Pricing Trends - AI server SSD demand is projected to exceed 45EB in 2024, with an average annual growth rate of over 60% expected in the coming years [7][9] - The average selling price (ASP) of NAND products has increased by approximately 15% due to AI demand, despite a 1% decline in shipment volume [9][10] - The enterprise SSD contract price has seen an accumulated increase of over 80% from Q4 2023 to Q3 2024 due to heightened AI server demand [7][9] Technological Innovations - SanDisk has introduced a new storage technology called High Bandwidth Flash (HBF), aimed at providing high bandwidth and capacity for AI inference applications [15][19] - HBF technology allows for parallel access to multiple high-capacity 3D NAND arrays, potentially offering a solution to compete with HBM in specific applications [15][19] - The first generation of HBF technology can provide up to 4TB of VRAM capacity, with plans for further enhancements [15][19] Future Outlook - NAND manufacturers are exploring various strategies to adapt to market changes, including enhancing storage density and dynamic capacity adjustments [22][23] - The introduction of HBF could signify a new opportunity for NAND manufacturers to capture a share of the AI market, although competition remains fierce [24][25] - The industry's recovery is evident, but ongoing challenges in the consumer electronics market continue to exert pressure on NAND manufacturers [11][24]

Core Viewpoint - NAND manufacturers are often overlooked in the AI boom, as their growth potential in AI applications is limited compared to DRAM, which is essential for AI computing needs [1][2][3] NAND Market Dynamics - AI computing primarily relies on DRAM due to its high bandwidth (up to 1TB/s) and low latency, while NAND's bandwidth (7-14GB/s) and higher latency make it less suitable for core AI tasks [1][2] - In AI training, NAND is mainly used for data storage rather than computation, leading to reduced demand once data is loaded into GPU DRAM [2] - The overall NAND market faces challenges such as easier capacity expansion and a larger number of competitors compared to DRAM, resulting in a more volatile market [3] Recent Market Trends - The NAND market experienced significant fluctuations in 2023, with a recovery in early 2024 driven by increased smartphone shipments and enterprise SSD demand due to AI server construction [5][6] - Global smartphone shipments rose by 8.2% year-on-year, and enterprise SSD procurement increased by 32% quarter-on-quarter, with North American cloud service providers accounting for 58% of global orders [5] - Strategic production cuts by major manufacturers like Samsung and Micron have led to a 17% increase in NAND Flash contract prices in Q1 2024 [5][6] Future Outlook - AI server SSD demand is expected to grow significantly, with predictions of over 45EB in procurement capacity in 2024 and an average annual growth rate exceeding 60% in the coming years [6][7] - The introduction of high-bandwidth flash (HBF) technology by SanDisk aims to provide a solution for AI inference applications, potentially positioning NAND to compete with HBM in the future [12][14] - The NAND industry is exploring various strategies, including enhancing storage density with QLC technology and dynamic capacity adjustments to address inventory issues [21][22] Competitive Landscape - The NAND market is characterized by a duality where enterprise demand is rising while consumer electronics face a downturn, leading to a stark contrast in performance across segments [8][9] - Companies are increasingly focusing on collaborative ecosystems and technology advancements to maintain competitiveness, with SK Hynix and Samsung leading in enterprise SSD markets [22][23]