"Korean Wave" is going crazy! Under the sweeping "Super Cycle of Storage," JPMorgan Chase loudly proclaims the Korean stock market soaring toward the "10,000-point era"

Bloomberg News has learned that Wall Street financial giant JPMorgan Chase has twice significantly raised its target level for South Korea’s benchmark stock index—the Kospi Composite Index—within less than a month. The core logic undoubtedly lies in the bull market storyline driven by the AI infrastructure frenzy amid the “storage chip supercycle” background, which is far from ending, coupled with corporate governance reforms led by South Korean President Lee Jae-myung and growth factors in the industrial sector. After the Asian trading session opened on Monday, the Kospi Composite Index surged over 5%, hitting a new all-time high, leading the Asia-Pacific markets strongly amid rising oil prices and escalating US-Iran tensions. Since the beginning of this year, the index has gained over 85%, leading the global markets, and is arguably the most疯狂 stock market in the world since 2026.

This largest commercial bank on Wall Street has raised the target for the South Korean Kospi Composite Index to 9,000 points and significantly increased its bullish scenario target to the epic milestone of 10k points, implying a potential upside of up to 33% from last Friday’s closing level. In comparison, the baseline and bullish targets set in late April were 7,000 and 8,500 points, respectively. As of press time, the Korean Kospi is hovering around 7,800 points.

EWY+7.61% MU+15.49% WDC+3.47% KS11+4.32% 000660+11.51% 005930+6.33%

Korean ETF - iShares MSCI

Follow

Analysis EWY

Included in our AI Select Strategy

·

View Strategy Details

190.20

▲+13.45(+7.61%)

Close·09/05·USD

189.05

▼-1.15(-0.60%)

After Hours·06:54:38

1 Day

1 Week

1 Month

6 Months

1 Year

5 Years

Max

Created with Highcharts 11.4.814:0015:0016:0017:0018:0019:00184186188190

Contents

EWY+7.61% MU+15.49% WDC+3.47% KS11+4.32% 000660+11.51% 005930+6.33%

Analysis EWY

Global stock market investors’ near-wild bullish enthusiasm around the Korean stock market all boils down to one core investment theme: an unprecedented AI boom-driven storage chip supercycle. Last Wednesday, overseas retail and institutional investors outside Korea bought over $2 billion worth of Korean stocks directly or indirectly through cross-border ETFs, just shy of the record set in October last year. The K-pop craze has spread from Seoul’s fashion circles to global financial markets.

Against the backdrop of this unprecedented AI infrastructure frenzy and the so-called “storage chip supercycle,” the two Korean-based giants—Samsung Electronics and SK Hynix—together accounting for nearly 50% of the weight in the Kospi Composite Index, are the most powerful engines attracting global capital. They are also the primary drivers behind Korea’s stock market repeatedly hitting new highs and significantly outperforming global equities. The Kospi index has already surpassed a wild 76% increase in 2026, which led global markets last year. But unlike the stark difference from the entire 2025, the 2026 rise exceeding 76% has occurred in less than five months since the start of the year.

Senior strategists from Wall Street are racing to raise their outlooks for the Korean stock market, mainly due to the exponential profit growth in the storage chip industry driven by the global AI boom. The Kospi surged 5.1% at one point on Monday, reaching an intraday high of 7,876.60 points, expanding its year-to-date gains to about 86%, cementing its position as the best-performing stock market globally. This JPMorgan Chase upgrade follows another Wall Street giant, Goldman Sachs, which last week raised its target for Kospi to 9,000 points, citing Korea’s strongest profit expansion momentum in Asia.

As shown in the chart above, the Korean Kospi index has risen over 80% this year—far outperforming the Philadelphia Semiconductor Index, known as the “global chip stock barometer.” Note: The index performance is standardized based on January 2, 2026.

As the Kospi’s rally continues, signs of overheating are increasingly evident. According to its 14-day Relative Strength Index (RSI), the index has been in overbought territory every trading day this month. However, including Morgan Stanley’s Mixo Das, the JPM strategists team wrote in a report that, despite the short-term technical overextension, “the key fundamentals of the market remain on track—optimism in the storage chip cycle, corporate governance reforms, and strong thematic growth.” “Under these unique conditions, we believe it is very appropriate to continue positioning for further upside rather than prematurely expecting the cycle to end.”

The strategists added that the next two years could mark a new phase of the storage chip industry entering a sustained boom driven by both average selling prices and record shipments. The two giants—SK Hynix and Samsung Electronics—whose market caps account for about 50% of the Korean stock benchmark, have driven this year’s roughly 70% increase.

Global capital is actively scrambling for Korean chip stocks. The iShares MSCI Korea ETF listed in the US has surged 95% this year, outperforming the US stock market and the Philadelphia Semiconductor Index. Investors focusing on the Hong Kong stock market are also actively buying leveraged chip ETFs linked to Korea’s single-chip sector, with the Hong Kong-listed 2x long SK Hynix ETF soaring 503% year-to-date, and the 2x long Samsung Electronics ETF rising 340%. Additionally, the China A-shares-listed Korea-China Semiconductor ETF has gained 117% this year.

Whether it’s Google’s enormous TPU AI computing clusters or Nvidia’s massive AI GPU clusters, both rely on fully integrated HBM storage systems that support AI chips. Coupled with the current tech giants’ accelerated construction or expansion of AI data centers, which require large-scale procurement of server-grade DDR5 storage and enterprise-level high-performance SSD/HDD; Samsung Electronics, SK Hynix, and Micron Technology are simultaneously caught in the three most critical storage domains: HBM, high-performance server DRAM (including DDR5/LPDDR5X), and high-end data center SSDs. They are the most direct beneficiaries of the “AI memory + storage stack,” reaping the “super dividends” of the AI infrastructure wave.

The Middle East conflict cannot dampen the “AI bull market” narrative! AI has fully ignited the “storage supercycle,” from HBM to NAND shortages.

This year, the Korean Kospi has ignored geopolitical negatives and surged about 85%, with Samsung Electronics’ market cap surpassing $1 trillion and SK Hynix’s stock price repeatedly hitting new highs. Essentially, this is not just a domestic Korean bull market but a global capital bet on the “AI-driven storage supercycle” that is far from over.

Research firm TrendForce predicted in early January that Q1 2026 contract prices for general-purpose DRAM would increase by 55%-60% quarter-over-quarter, and NAND Flash by 33%-38%. However, by early February, due to worsening AI and data center demand and a global storage supply-demand imbalance, the market research firm TrendForce revised the Q1 increases sharply upward to 90%-95% for general DRAM and 55%-60% for NAND Flash, noting that PC DRAM could increase over 100% quarter-over-quarter, server DRAM about 90%, and enterprise SSDs 53%-58%.

Regarding the price surge of DRAM/NAND storage chips, Goldman Sachs’ latest judgment is that the storage price increases in 2026 will far exceed previous optimistic expectations. Goldman recently raised its forecast for DRAM price increases from about 150% to 250%-280%, and NAND prices from about 100% to 200%-250%. In other words, Goldman believes this is not just an ordinary inventory correction cycle but an “super cycle” driven by unprecedented demand fueled by AI computing power, with HBM increasingly occupying capacity due to complex manufacturing and packaging processes, and limited supply elasticity for general DRAM/NAND.

GPU is responsible for generating intelligence, HBM/DRAM for高速 data feeding, enterprise SSDs for hot data and caching, and HDDs for long-term storage of massive cold/warm data. Goldman believes that the AI computing race led by cloud giants is transforming storage chips from cyclical commodities into strategic scarcity assets. The price increases of DRAM/NAND in 2026 are not the end but may mark the initial phase of a super cycle.

As Micron’s senior vice president and general manager of data center business Jeremy Werner revealed in a recent interview, from the underlying logic of AI data center data flow processing, the driving force behind this cycle is not simply “AI needs more compute chips,” but rather the era of AI inference led by agents like Claude Cowork and OpenClaw is turning memory/storage from supporting components into system bottlenecks.

AI training relies heavily on large-scale parallel computing, but inference—especially with long contexts, multi-turn dialogues, and agentic AI workflows—requires continuous storage of KV caches, context states, and intermediate results. When memory/storage is insufficient, models must recompute historical states, reducing GPU utilization and increasing token generation costs. Therefore, HBM, DDR5, LPDDR, enterprise SSDs, and even HDDs/data lakes are forming an “AI memory chain” from GPU proximity to remote storage, determining throughput, latency, concurrency, and token economy of AI systems. This explains why storage and data storage stocks like Micron, Samsung, SK Hynix, SanDisk, Western Digital, and others are soaring in tandem: demand is not only concentrated on HBM but spills over along the entire AI server architecture into DRAM, NAND, SSD, and HDD supply chains.

More critically, AI CPUs are opening a second demand curve. In the past, AI compute power was almost synonymous with GPU + HBM, but as inference workloads become more complex, CPUs are upgrading from “GPU sidekick” to “AI orchestrator,” managing multiple agents, contexts, and workflows, significantly boosting server DDR5 and data center SSD demand. Meanwhile, HBM capacity is heavily locked by AI GPUs, and general-purpose DRAM capacity is squeezed, causing divergence in DDR5 and DDR4 prices, with shortages spilling from high-end HBM to broader DRAM/NAND supply chains. TrendForce also cited Micron’s CEO’s latest view that both traditional and AI server demands are strong but constrained by DRAM and NAND supply tightness; Samsung and SK Hynix recently warned that AI-driven storage shortages could persist until 2028 or longer.

Feel the shock of the “storage supercycle”! Cloud giants are not only competing for HBM but also actively helping SK Hynix buy lithography equipment to expand capacity.

Star analyst Ben Reitzes and his team at Melius recently published a report stating that the AI boom will sustain strong growth in storage chip demand into the end of this decade (2030). According to market research firm Counterpoint Research, the storage market has entered a “super bull market” or “super cycle,” with current supply, demand, and pricing far surpassing the peaks seen during the 2018 cloud computing boom.

As Anthropic’s Claude Cowork and autonomous AI agent tools like OpenClaw explode in 2026, this wave of AI agents is rapidly sweeping the globe. The bottleneck in AI compute architecture is shifting from matrix multiplication throughput (GPU) to “AI agent-driven full-stack AI systems,” with data center CPUs and storage chips likely becoming the biggest winners in this shift. In this AI narrative transformation, data center CPUs and storage chips may be the greatest beneficiaries. In other words, the AI compute bull market is expanding from “AI GPU/ASIC compute systems” to central processors and “data storage foundations.”

Media reports citing insiders say that SK Hynix, the super leader in HBM storage, is facing “unusual” active competition from major global tech companies—Microsoft, Google, and Amazon, the world’s largest cloud giants, are unprecedentedly proposing large-scale investments in new production lines and planning to personally fund the purchase of increasingly expensive chip manufacturing equipment, including ASML lithography machines, advanced HAR etching, and thin-film deposition equipment, to expand capacity. Their goal is to lock in as much capacity and supply as possible while competing to ensure HBM, DRAM, and NAND chip supplies.

This approach of bidding and investing in capacity is unprecedented in the global storage chip industry, highlighting the severe shortage of these components worldwide. Amid the unprecedented surge in AI-driven compute infrastructure demand, storage chip manufacturers are generally struggling to keep pace with exponential expansion needs.

Three insiders said another proposal involves customers providing huge funding support for semiconductor manufacturing equipment procurement—such as ASML’s extreme ultraviolet (EUV) lithography machines or even more expensive high-NA lithography tools; these are used for printing circuits, etching, deposition, CMP, and other cutting-edge chip manufacturing processes, worth billions of dollars.

However, two of these insiders noted that the Korean chip manufacturer has ample cash flow and is cautious about accepting customer financial and investment commitments, because such deals could make it dependent on specific buyers and might further require selling storage chips at below-market prices in exchange for longer-term, more stable revenue.

Storage chip manufacturers have recently emphasized that multi-year contracts will help smooth demand volatility and reduce the huge investment risks in this cyclical industry, which often requires billions of dollars to significantly expand capacity.

The unprecedented AI infrastructure wave and the storage supercycle have pushed semiconductors into a new stage—more “material-intensive, process-controlled, and advanced packaging forward”—with the combination of 3D logic structures, new materials, HBM stacking and interconnect upgrades, and advanced packaging like CoWoS/hybrid bonding transforming system performance into manufacturing complexity. These forces collectively increase the value density of key steps like deposition, etching, CMP, advanced packaging, and metrology, and more clearly rewrite the demand for semiconductor equipment from “cyclical fluctuations” to “structural large-scale expansion.”

The most critical semiconductor equipment base for storage chip expansion includes not only ASML lithography machines but also high-aspect-ratio (HAR) etching/deposition, CMP, metrology/inspection, and hybrid bonding equipment—expensive high-end tools needed for HBM, DRAM, and NAND. However, these two Korean storage chip suppliers are very cautious in how they allocate scarce capacity to avoid antitrust scrutiny or being perceived as favoring specific customers. “They don’t want to ‘bet on a single horse’ in the AI race and end up making the wrong choice,” said one insider.