Kioxia Shares Sink After Outlook Shows Demand Leveling Off—And the AI Memory Trade Just Got a Reality Check

 

Kioxia Shares Sink After Outlook Shows Demand Leveling Off—And the AI Memory Trade Just Got a Reality Check

**By [moonlight editory team] | November 3, 2026 | 11-Minute Read**

---

## The Short Version (For the Busy American)

Kioxia Holdings just delivered a brutal reminder that even the hottest AI trade can stumble on something as mundane as a shipping schedule.

The Japanese memory chipmaker's shares **plunged as much as 17% on Monday**, closing down roughly **15.8%** at 11,345 yen. At one point, the stock hit its lowest level in six months. It's now down nearly **70% from its June peak**, when Kioxia briefly became Japan's most valuable listed company on the back of the AI memory boom .

The trigger wasn't a bad quarter. Kioxia's September-quarter results were stunning: revenue hit **¥2.48 trillion**, up **453% year-over-year**, and operating profit surged more than **20-fold** to **¥1.99 trillion** . The problem was the guidance. Kioxia told investors to expect operating profit of just **¥2 trillion** for the December quarter—barely above the prior quarter and well below the **¥2.35 trillion** analysts expected .

**The punchline:** Kioxia says the slowdown is a timing issue, not a demand problem. But the market isn't buying it. And if the NAND price surge is actually peaking, the entire AI memory trade—from SK Hynix to Micron to SanDisk—could be next.

---

## What Actually Happened: The Guidance Miss

Let me walk you through the numbers, because the contrast between the quarter and the outlook is what spooked investors.

### The Blowout Quarter

Kioxia's September quarter was one for the record books:

| Metric | Q2 FY2026 | Change |

|--------|-----------|--------|

| **Revenue** | ¥2.48 trillion | +453% YoY |

| **Operating Profit** | ¥1.99 trillion | +2,181% YoY |

| **Net Income** | ¥1.36 trillion | +61% QoQ |

| **NAND ASP Growth** | +20-25% QoQ | Down from +70% prior quarter |

*Source: Company earnings, Goldman Sachs analysis* 

The NAND price surge was the engine. Average selling prices rose **20-25%** in the quarter—a massive increase, but a sharp deceleration from the **70%** jump in the previous quarter . That deceleration is the number that matters.

### The Disappointing Guidance

Kioxia projected December-quarter revenue of **¥2.5 trillion**—up just **0.8% sequentially**—and operating profit of **¥2 trillion**, essentially flat from the prior quarter .

The company blamed **shipment timing**. Some orders originally scheduled for the December quarter were pulled forward into September at customer request. Others were pushed into the March quarter because enterprise storage products require longer production lead times .

CFO **Yoshihiko Kawamura** insisted the company expects to "return to rapid growth in the March quarter" and that demand from AI data center operators remains robust .

### The Market's Skepticism

Investors weren't reassured. UBS Securities Japan's research head, **Kenji Yasui**, pressed the executives on the call: **"You say supply and demand will be extremely tight heading into the next fiscal year, but that the price outlook is uncertain. From a market perspective, you would normally think prices would continue to rise for a while. Are you just being conservative?"** 

That question captures the problem. Kioxia says demand is so strong that its **2027 capacity is sold out** and more than half of its **2028 output** is already allocated through long-term agreements. Some customers are even negotiating for **2029 and 2030** . Yet the company is guiding for flat profits.

Either the guidance is overly conservative, or something in the pricing picture is changing. The market is betting on the latter.

---

## The Human Touch: What This Means for the AI Trade

Let me put this in terms that actually connect to your portfolio and your understanding of the AI boom.

### The Memory Trade Was the Purest AI Play

For the past year, memory chips were the simplest way to bet on AI. Nvidia makes the brains. But those brains need memory—**high-bandwidth memory (HBM)** for training, and **NAND flash** for storage. Companies like Kioxia, SK Hynix, Samsung, and Micron became the picks-and-shovels play for the AI gold rush.

Kioxia's stock **rose more than 750%** in the first half of 2026. It was a pure NAND play, and NAND prices were exploding because AI data centers needed storage .

The guidance miss raises a question that's been lurking for months: **when does the price surge peak?**

### The Deceleration Signal

Here's the number that should concern anyone invested in memory stocks. Kioxia's NAND ASP growth went from **70% in Q2 to 20-25% in Q3**, and the company is guiding for **low-single-digit growth** in Q4 .

Prices are still rising. But the **rate of increase** is collapsing. And for a stock that's priced for accelerating growth, a decelerating growth rate is enough to trigger a selloff.

Analysts at Goldman Sachs noted that the Q4 guidance miss may partly reflect **shipment timing and a higher base** compared to Korean competitors. They maintained their Buy ratings on **SK Hynix and Samsung**, arguing that NAND supply-demand tightness will persist through 2027 and support further price increases .

But the market's reaction shows that investors are no longer willing to give memory companies the benefit of the doubt.

### The DRAM Advantage

There's a structural reason why Kioxia might be more vulnerable than its Korean rivals. **Kioxia is a NAND-only company**. SK Hynix and Samsung sell both **DRAM** and NAND. In an AI computer, DRAM is just as critical as NAND—it's the working memory that feeds the processor.

That means Samsung and SK Hynix have **more bargaining power** with customers. They can bundle DRAM and NAND, negotiate better terms, and capture more value. Kioxia can only sell one product .

When the pricing cycle turns, the NAND-only player feels it first.

### The Leverage Problem

Adding to Monday's selloff: Kioxia has the **highest concentration of leveraged long positions** among all companies on the Tokyo Stock Exchange, according to analysis cited by Eastmoney. When a negative catalyst hits, margin traders are forced to sell, amplifying the decline .

That's a technical factor, not a fundamental one. But it explains why Kioxia's stock fell so much harder than the broader market.

---

## Frequently Asked Questions

### Why did Kioxia stock fall so much?

Because its **December-quarter guidance missed expectations**. Kioxia projected operating profit of **¥2 trillion**, below the **¥2.35 trillion** consensus and essentially flat from the prior quarter. Investors interpreted the guidance as a sign that the AI-driven NAND price surge is leveling off .

### Did Kioxia have a bad quarter?

**No.** The September quarter was record-breaking: revenue up **453%** and operating profit up more than **20-fold**. The problem was the outlook, not the results .

### Why is the guidance so weak if demand is strong?

Kioxia blames **shipment timing**. Some orders were pulled forward into September; others were pushed into March because enterprise storage takes longer to produce. The company says demand from AI data centers remains robust and expects a return to rapid growth next quarter .

### Is the NAND price surge ending?

**It's decelerating.** NAND ASP growth slowed from **70% in Q2 to 20-25% in Q3**, and Kioxia is guiding for low-single-digit growth in Q4. Prices are still rising, but much more slowly. Whether that's a temporary pause or the start of a peak is the debate .

### How does this affect other memory stocks?

**It's a warning sign.** SanDisk fell **13.5%** and Micron dropped **5.3%** on Friday after Kioxia's outlook was released . However, U.S. memory stocks actually **rallied in premarket trading Monday**, suggesting some investors see Kioxia's issues as company-specific rather than sector-wide .

### What does Goldman Sachs say?

Goldman maintained **Buy** ratings on **SK Hynix and Samsung**, arguing that NAND supply-demand tightness will persist through **2027** and support price increases of **34% and 30%** respectively. The firm sees Kioxia's guidance miss as partly a timing and base-effect issue .

### What is Kioxia's capacity situation?

Kioxia says its **2027 capacity is sold out**, more than half of its **2028 output** is allocated through long-term agreements, and customers are negotiating for **2029-2030** supply. That's a sign of strong long-term demand—but it hasn't stopped the stock from falling .

### Is this a buying opportunity?

I can't give personalized investment advice. What I can tell you is that **Kioxia is down 70% from its peak** and trades at a much lower valuation than it did in June. The long-term demand picture for AI memory remains strong. But the deceleration in pricing is a real signal, and the stock's heavy leverage exposure makes it volatile. The risk/reward depends on your view of the AI memory cycle.

---

## The Path Forward: What to Watch

### The March Quarter Guidance

Kioxia says it will "return to rapid growth" in the March quarter. If it delivers, the stock could recover. If it misses again, the "timing issue" narrative falls apart.

### NAND Pricing Data

Watch the monthly NAND contract price data from TrendForce and other research firms. If prices keep rising at a healthy pace, Kioxia's guidance was too conservative. If they flatten or fall, the memory trade is in trouble.

### Samsung and SK Hynix Earnings

The Korean memory giants report in the coming weeks. Their guidance will tell us whether Kioxia's slowdown is company-specific or industry-wide. If Samsung and SK Hynix also guide conservatively, the entire memory trade faces a repricing.

### The AI Capex Picture

Alphabet's free cash flow turned negative. Amazon is spending **$220 billion** on capex. Meta is raising **$95 billion** in equity. If the hyperscalers keep spending, memory demand stays strong. If they pull back, the memory trade loses its foundation.

---

## Conclusion: The First Crack in the Memory Boom

Here's what I want you to take away from Kioxia's collapse.

The AI memory trade was the purest way to bet on the AI boom. No complex software. No regulatory uncertainty. Just supply and demand: AI data centers need memory, memory prices go up, memory stocks go up.

Kioxia's guidance just showed that the "prices go up" part isn't guaranteed. The surge is decelerating. The market is repricing. And the stock that rose 750% in six months just gave back a huge chunk of those gains in a single day.

The company says the demand is still there. The capacity is sold out. The long-term agreements are signed. But the market is asking a simple question: **if demand is so strong, why is your profit guidance flat?**

Kioxia's answer—"timing"—may be true. But investors have heard "timing" before. And they're not sure they believe it this time.

The memory boom isn't over. But the easy money has been made.

---

## Disclaimer

This article is for informational purposes only and does not constitute financial advice, investment advice, or a recommendation to buy or sell any securities. The author is not a licensed financial advisor. All investments involve risk, including the potential loss of principal. Past performance does not guarantee future results. Readers should conduct their own research and consult with a qualified financial professional before making any investment decisions. Data and figures cited are accurate as of the date of publication but may change. The author holds no position in any securities mentioned.

---

**Tags:** #Kioxia #NAND #MemoryChips #AIStocks #StockMarketToday #FinancialNews #MarketAnalysis #Investing #SemiconductorStocks #Samsung #SKHynix #Micron #SanDisk #GoldmanSachs #AI Trade #EarningsSeason #StockMarket #JapanStocks #Nikkei #TechStocks #InvestmentStrategy #RiskManagement #MemoryCrisis #HBM #DRAM #AIInfrastructure #DataCenters #MarketVolatility #EarningsMiss #GuidanceCut