Storage surges again! AI inference and multimodal drives a data explosion, making hard disk and flash memory manufacturers the biggest beneficiaries.

Storage surges again! AI inference and multimodal drives a data explosion, making hard disk and flash memory manufacturers the biggest beneficiaries.

```

As the AI wave shifts from the training phase to large-scale deployment in inference applications, the storage sector—seen as the "working memory" of AI—is undergoing an unprecedented revaluation.

On Tuesday, storage concept stocks in the U.S. market surged collectively, with SanDisk skyrocketing 27.56%, marking its best single-day performance since February. Western Digital and Seagate Technology followed closely, rising 16.77% and 14.00% respectively.

The direct catalyst for this rally came from Nvidia CEO Jensen Huang’s speech at CES. He bluntly stated: “As far as storage is concerned, this is currently a completely untapped market. This is a market that never existed, and it’s very likely to become the largest storage market in the world, essentially bearing the world’s AI working memory.” At the same time, Nvidia showcased a new storage platform at CES optimized for agent AI inference, promising to improve energy efficiency fivefold over traditional platforms.

This surge is no coincidence, but rather a profound reassessment by the market of the current phase of AI development. Bank of America analyst Wamsi Mohan pointed out in his latest report that 2026 will be a turning point for enterprise-level and edge AI. With the proliferation of multimodal AI (including text, images, and video), the amount of data generated will grow exponentially, driving a sustained hardware spending cycle.

Bank of America believes that the theme of AI investment is shifting from capital expenditure-driven model training to an AI inference phase focused on return on investment (ROI). This transition will make storage, edge devices, and network connectivity manufacturers the new beneficiaries following GPUs.

AI Inference and Multimodality: The Real Drivers of Data Explosion

The core logic of the storage sector’s revaluation lies in the nature of AI workloads transitioning.

According to IDC data, global annual data creation is expected to soar from 173 ZB in 2024 to 527 ZB in 2029—over twice the growth in five years, with a compound annual growth rate of about 25%. The data explosion from AI models is placing unprecedented demands on storage capacity and speed.

Analyst Wamsi Mohan emphasized in the report, “As we look toward 2026 and beyond, we expect AI inference to dominate.” Enterprises need to retain more and more data for training, analytics, and compliance purposes, resulting in a “synchronous surge” in storage demand.

Especially with the rise of multimodal generative AI. Unlike early text-based AI, the spread of multimodal AI means systems must process and generate unstructured data like images, video, and audio. These data types are not only massive, but require frequent read and write operations to support closed loops of “monitor/verify/reintegrate,” making storage not just a passive archiving tool, but an essential active participant in AI computational workflows.

Hard Drive Renaissance: Opportunities for Seagate and Western Digital

In mass data storage, hard disk drives (HDDs) continue to hold an irreplaceable position due to cost advantage and capacity density. Bank of America believes the increasing demand from multimodal AI and inference will directly drive HDD shipments upward and prompt clients to shift to higher-capacity drives, thereby increasing the per-terabyte value for Seagate (STX) and Western Digital (WDC).

  • Mass Storage Demand (HDD): Seagate (STX) and Western Digital (WDC) are major beneficiaries. Content generated by multimodal AI (such as generative video, surveillance logs, synthesized images) requires low-cost, high-capacity storage media. Seagate’s HAMR (Heat-Assisted Magnetic Recording) technology and Western Digital’s UltraSMR technology are designed to meet the extreme demands for single-drive capacity and energy efficiency.
  • High-performance Cache Demand (NAND): Inference is not “read-only.” Modern AI systems need to store prompts, feedback labels, security logs, and vector databases for RAG (Retrieval Augmented Generation). This requires numerous random I/O and write operations, directly boosting demand for high-performance enterprise SSDs.

Rise of Edge AI: The Battlefield for SanDisk and High-Performance Flash Memory

Jensen Huang mentioned at CES: “Storage is currently a completely undeveloped market… this may become the world's largest storage market, essentially bearing the world’s AI working memory.”

This view aligns with Bank of America's analysis. Beyond data centers, AI is quickly permeating edge devices. Edge AI—running AI directly on devices like smartphones, PCs, vehicles, and drones—will be another huge area of growth.

  • SanDisk’s Opportunity: Edge AI requires extremely low latency and high reliability. Devices must locally process high-definition video streams, sensor data, and real-time decisions, driving the transition of storage media from low-end to high-performance UFS and NVMe interfaces. As a leader in embedded and removable flash memory, SanDisk will benefit from rising device-level storage capacity and product structure upgrades (such as automotive-grade UFS 4.1).
  • Resurgence of Terminal Manufacturers: Apple (AAPL) is seen as the “ultimate edge AI player”—its vast device base and self-developed chips (M-series/A-series) enable local inference with privacy protection and ultra-fast response. With its massive installed base, custom Apple Silicon chips, and pursuit of privacy, Apple can deliver faster and more secure experiences through on-device AI. Integrating third-party models like Gemini with Siri may further increase ecosystem stickiness. Meanwhile, Dell (DELL) and HP (HPQ) will also benefit from enterprise demand for “AI PCs,” with Gartner forecasting that AI PCs will account for 43% of all PC shipments in 2025.

Price Increases and Market Outlook

Amid booming demand, supply constraints are supporting the rally in stock prices. According to Bloomberg Industry Research analyst Jake Silverman, as demand for AI training and inference grows, storage supply is tightening and prices are soaring. Earlier this week, Korea Economic Daily reported that Samsung Electronics and SK Hynix are seeking to hike server DRAM prices by 60% to 70% in Q1 compared to Q4 last year.

Bank of America concludes that hardware spending as a share of IT industry revenues has risen year by year since 2022, and the industry is entering a “hardware renaissance era.” In this cycle, not only Nvidia—which supplies compute power—but also storage suppliers (providing “memory”), and Amphenol and Corning (providing connectivity), will be central beneficiaries of the AI inference wave.

 

~~~~~~~~~~~~~~~~~~~~~~~~

The wonderful content above comes from Chasing the Wind Trading Desk.

For more in-depth insights, including real-time analysis and frontline research, please join [Chasing the Wind Trading Desk • Annual Membership]

Risk Warning and DisclaimerThe market carries risks, and investments should be made cautiously. This article does not constitute personal investment advice and does not take into account the specific investment objectives, financial situations, or needs of any individual user. Users should consider whether any opinions, views, or conclusions in this article are suitable for their particular circumstances. Investments made based on this article are at your own risk. ```