● AI-Growth-Explosion, Dollar-Surge, Mega-Rally

In the High-Rate New Normal, What Wall Street Means by the “AI Growth Cycle” and the Strong-Dollar Scenario

The core issue is not simply that “rates are high and equities are risky.”

The main argument advanced by Cathie Wood and ARK, prominent investors in innovative growth stocks, is more ambitious.

U.S. Treasury yields may rise further and the Federal Funds rate may remain elevated, but if the backdrop is AI infrastructure investment and a productivity revolution rather than inflation, equities could strengthen further.

This framework connects the strong-dollar thesis, global capital flows into U.S. equities, Nasdaq strategy, long-duration bond risk, and gold price outlook.

In particular, this report addresses a point often missed by mainstream coverage: AI is not just a theme, but a payment system drawing capital into the United States.

1. The market’s central question: Is high interest-rate policy truly negative for equities?

Rates remain the market’s key variable.

The U.S. 10-year Treasury yield has moved to around 5%, and the market also discusses the possibility of levels above 5.1% or 5.2%.

Under conventional assumptions, higher rates pressure growth and technology stocks.

Higher discount rates reduce the present value of future earnings.

For that reason, AI stocks, biotech stocks, and innovative growth companies with more future than current earnings have typically underperformed in rising-rate periods.

However, ARK’s Cathie Wood offers a different interpretation.

She views the current high-rate environment not as a warning signal but as a growth signal.

If rates are rising because the U.S. economy is strong and AI investment is lifting GDP growth, the environment may be less concerning.

In other words, it is necessary to distinguish between bad rate increases and good rate increases.

• Bad rate increases: higher rates driven by inflation, surging energy prices, fiscal stress, or credit risk
• Good rate increases: higher rates driven by productivity gains, economic growth, and stronger corporate investment

Cathie Wood’s view is closer to the latter.

Her argument is that if the AI revolution raises the U.S. economy’s growth rate, higher Treasury yields do not necessarily imply a market downturn.

2. ARK’s view on the U.S. 10-year Treasury: 5% is not historically abnormal

ARK supports its view with long-term yield history.

Over roughly 130 years, the U.S. 10-year Treasury yield has generally traded within a 3% to 6% range.

By that standard, the near-zero-rate environment that followed COVID-19 was unusually low rather than the current 5% level being unusually high.

From that perspective, 5% is within the historical range.

That said, the market experience is different.

For investors accustomed to near-zero rates in the early 2020s, 5% is a meaningful shock.

This is especially true for investors in long-duration Treasury ETFs such as TLT, where rising yields translate directly into lower prices.

ARK goes further.

If the AI revolution materially raises economic growth, long-term yields could move above 6% and potentially toward 7% to 8%.

This is a controversial scenario.

Yields at 7% to 8% would be a major negative for long-duration bonds and for highly leveraged companies.

ARK’s view is that markets could absorb higher yields if growth also rises meaningfully.

3. Cathie Wood’s core assumption: The AI revolution may be larger than the Industrial Revolution

ARK’s willingness to tolerate higher rates rests on one premise.

It sees the current technological cycle as more powerful than previous industrial revolutions.

Past industrial revolutions were driven by electricity, telephony, and the internal combustion engine.

ARK argues that five innovation platforms are now advancing simultaneously:

• Artificial intelligence
• Robotics
• Energy storage
• Genomics and biotechnology
• Blockchain

The key point is that these platforms do not operate in isolation.

AI becomes the brain of robotics, energy storage extends the operating range of EVs and robots, genomics accelerates drug discovery through AI, and blockchain may reshape digital finance and data trust frameworks.

ARK argues that this convergence could produce not just technological progress, but a broad productivity revolution across the economy.

On that basis, ARK sees U.S. economic growth potentially rising from around 3% to roughly 7% over the next 3 to 5 years.

It also cites Elon Musk’s comments on the possibility of 20% growth, arguing that conventional economic models may not fully capture the next cycle.

This is an aggressive forecast.

However, the debate around AI infrastructure spending, data center construction, semiconductor demand, and cloud expenditure contributing to GDP is increasingly prominent on Wall Street.

4. Satya Nadella’s AI agent view: The speed and scale are different

This discussion is not limited to Cathie Wood.

Microsoft CEO Satya Nadella has made similar remarks.

He has said the AI agent market could be faster, larger, and more powerful than the PC era or the cloud era.

Microsoft is a company built on enterprise customers.

As a result, it is well positioned to observe whether companies are actually realizing productivity gains from AI adoption.

If AI extends beyond search and chat into document generation, coding, accounting, customer support, sales automation, and data analysis, corporate cost structures will change.

The key issue is the speed of adoption.

• The electricity revolution took decades to diffuse widely.
• The PC revolution also required more than a decade to scale.
• Cloud migration took substantial time.
• AI agents, by contrast, may spread across enterprise workflows within only a few years.

If that pace holds, AI infrastructure spending may become a core driver of U.S. growth.

That would support investment across semiconductors, data centers, power grids, cooling systems, cloud platforms, software, security, and automation.

5. ARK’s most important variable: Cost declines drive market expansion

Cathie Wood consistently emphasizes cost.

She defines the essence of technological innovation as cost reduction.

Electricity enabled mass adoption of appliances, falling lithium-ion battery prices expanded the EV market, and lower computing costs supported the spread of PCs and smartphones.

AI follows the same pattern.

If training and inference costs decline rapidly, AI becomes accessible to more companies and individuals.

Capabilities that once cost tens or hundreds of thousands of won can now be accessed through a monthly $20 subscription or even at no cost.

As OpenAI, Anthropic, Google, and Meta compete, AI usage costs continue to fall.

Lower costs increase usage.

Higher usage drives more AI infrastructure investment.

That cycle can benefit Nvidia, AMD, Micron, Broadcom, TSMC, cloud providers, and power infrastructure companies.

Biotechnology is following a similar pattern.

Genetic analysis costs have fallen sharply, and AI is helping researchers analyze protein structures and drug candidates more quickly than before.

ARK characterizes this as a productivity revolution.

6. The key point often missed: AI is a subscription engine that pulls capital into the United States

This is the most important issue.

Many reports focus only on AI semiconductors, data centers, and Nvidia earnings.

But at a deeper level, AI is creating a payment structure that channels global capital into the United States.

For Korean investors, the effect is easy to observe.

In the past, the main ways to spend on U.S. companies were buying iPhones, subscribing to Netflix, or using Google Cloud and Apple iCloud.

Now that list also includes ChatGPT, Claude, Gemini, Grok, GitHub Copilot, and Microsoft 365 Copilot.

This applies not only to individuals.

At the enterprise level, company-wide deployment creates recurring monthly subscription payments.

As firms around the world adopt AI tools, monthly dollar-denominated payments flow to U.S. large-cap technology companies.

This is not only a technology trend; it may also affect the balance of payments and exchange rates.

Regardless of which company ultimately wins the AI market, the leading candidates are mostly U.S.-based.

• OpenAI is a U.S. company.
• Anthropic is a U.S. company.
• Google Gemini is offered by a U.S. company.
• Microsoft Copilot is offered by a U.S. company.
• xAI’s Grok is also centered in the U.S. ecosystem.

In other words, broader AI adoption supports revenue growth for U.S. companies and can increase demand for dollars.

This is the connection to the strong-dollar thesis.

AI could function not only as an earnings driver but also as a mechanism reinforcing dollar dominance.

7. ARK’s strong-dollar view: Could a 1980s-style super dollar return?

Cathie Wood has recently taken a strong position on the dollar.

Markets often discuss dollar weakness, but ARK argues that the dollar’s relative value is in fact strengthening.

Standard dollar indices are usually based on the euro, yen, and pound.

ARK argues that the real trade-weighted dollar against 26 U.S. trading partners is a better measure.

On that basis, the dollar’s long-term trend remains upward.

ARK’s support for a strong dollar rests on three factors.

• Pro-business U.S. policy
• Monetary policy that preserves dollar credibility
• Dominant U.S.-centered technological innovation

The third factor is the most important.

AI, space technology, cloud, semiconductor design, biotechnology, and software platforms are concentrated in the United States.

If unlisted innovation companies such as SpaceX and OpenAI eventually enter public markets, global capital could continue to flow toward U.S. equities.

That would support dollar strength.

For Korean investors, however, a strong dollar has mixed implications.

It may create foreign-exchange gains for holders of U.S. assets, but it also increases local-currency costs for living expenses and imported goods.

If the won weakens materially, return calculations become more complex.

8. Gold price outlook: Is gold weaker if the dollar is stronger?

ARK is relatively cautious on gold in the context of a stronger dollar.

Gold typically benefits when confidence in the dollar weakens, inflation rises, or geopolitical risk increases.

However, if dollar credibility improves and U.S. monetary policy is viewed as stable, gold’s appeal may weaken.

A stronger dollar often pressures gold prices in international markets.

That said, Korean investors should consider one additional factor.

Even if gold prices fall in dollar terms, a weaker won can push gold prices higher in local-currency terms.

In other words, dollar-denominated gold and won-denominated gold may move differently.

For Korean investors, the relevant variables are not only the gold price itself but also the USD/KRW exchange rate, real rates, the U.S. fiscal deficit, and central bank gold purchases.

9. The most important warning for long-duration bond investors: TLT may not be done falling

The most practical warning in this discussion concerns long-duration U.S. bonds.

If ARK’s scenario proves correct and U.S. growth rises while long-term yields continue to increase, long-duration bond prices could fall further.

This is especially relevant for TLT and other long-maturity bond ETFs, which are highly sensitive to interest-rate changes.

Many investors assume that because yields have already risen substantially, bonds are now attractive.

The problem is that it remains unclear whether yields have peaked.

The result for bond investors will depend entirely on whether the yield rise reflects a final pre-recession move or a new, higher nominal-growth regime driven by AI productivity.

For that reason, long-duration bonds should not be approached simply because prices are lower; the direction of rates and the investment horizon matter.

10. ARK’s portfolio direction: Moving capital toward innovation assets

ARK’s conclusion is clear.

Rather than remaining in cash, gold, or long-duration bonds to avoid high rates, investors should reallocate toward innovation assets.

ARK’s preferred areas include:

• Tesla and autonomous driving
• Robotics and automation
• Space technology and the SpaceX ecosystem
• AI software and cloud infrastructure
• Biotechnology and genomics
• Blockchain and digital finance
• Coinbase, Robinhood, and stablecoin-related companies

There are also conflicts of interest.

ARK manages ETFs focused on innovative growth stocks.

It is therefore natural for the firm to present arguments favorable to innovation equities.

There is also a contradiction in its message.

While Cathie Wood says highly leveraged companies should be treated cautiously in a high-rate environment, many of ARK’s holdings are loss-making or dependent on external financing.

Investors should view this critically.

Even if the AI growth narrative is correct, not every innovation company will survive.

In a high-rate environment, weak balance sheets can be a decisive disadvantage.

11. Investment interpretation: Nasdaq accumulation may be more practical than picking individual winners

Even if one agrees with ARK’s direction, not every investor needs to select Tesla, biotechnology, blockchain, or robotics stocks individually.

For investors with limited information, a diversified approach through the Nasdaq 100 or the S&P 500 may be more practical.

If the AI revolution genuinely lifts the U.S. economy, the benefits are unlikely to be limited to one or two companies.

The impact could extend across semiconductors, cloud, software, platforms, power infrastructure, data centers, and automation firms.

In that case, accumulating U.S. equity indices or Nasdaq exposure may be more effective than assuming single-stock risk.

For long-term investors, participating in the broader U.S. innovation ecosystem may matter more than identifying the exact winner.

Still, in a high-valuation environment, phased buying and cash management remain important.

12. Key indicators to monitor

To assess whether this scenario is unfolding, several indicators should be watched closely.

• Whether the U.S. 10-year yield stabilizes above 5%
• Whether expectations for Federal Reserve cuts continue to fade
• Whether AI infrastructure spending is reflected in GDP and corporate earnings
• Whether earnings from Micron, Nvidia, and Broadcom remain strong
• Whether power demand and power infrastructure investment continue to expand
• Whether revenues at OpenAI, Anthropic, and xAI continue to grow
• Whether the USD/KRW exchange rate reflects continued dollar strength
• Whether gold remains resilient despite dollar strength
• Whether long-duration bond ETFs continue to face downside pressure

The most important question is whether AI translates into measurable productivity gains.

If AI becomes only a cost burden and investment bubble, downside risk increases.

If companies use AI to reduce labor, operating, and R&D costs while increasing revenue, equities may remain supported even in a high-rate environment.

13. The most important alternative scenario

ARK’s outlook is compelling, but the opposite scenario must also be considered.

First, higher yields may reflect fiscal deficits and Treasury supply pressures rather than growth.

If U.S. government debt continues to rise and Treasury issuance expands, investors may demand higher yields.

In that case, rising rates would not be positive for equities.

Second, AI spending may become overheated and turn into a bubble.

Railroads transformed the economy, but many early investors still suffered when the bubble burst.

AI may also prove correct over the long term while prices overshoot in the short term.

Third, an excessively strong dollar could strain emerging markets and global liquidity.

A sharp dollar rise raises debt burdens outside the U.S. and increases global volatility.

Fourth, many innovation companies remain vulnerable to high rates.

Firms with low revenue and weak cash flow are directly exposed to higher financing costs.

For that reason, investors should examine balance sheets and cash flow carefully even when investing in AI.

14. Conclusion: In the high-rate era, growth quality matters more than the rate level itself

The most important message from this Wall Street view is not that high rates are automatically negative.

What matters is why rates are rising.

If rates rise because of inflation and fiscal stress, that is a warning sign.

But if they rise because of AI infrastructure investment, productivity gains, stronger corporate earnings, and higher U.S. growth, the interpretation changes.

ARK is framing the current market in the latter way.

For that reason, it remains constructive on U.S. equities, innovative growth stocks, AI-related names, and dollar strength.

By contrast, it is cautious on long-duration bonds and gold.

However, this is a highly optimistic view, and investors should also recognize that ARK’s own incentives may shape the message.

For individual investors, the most practical approach is not to bet on extreme forecasts, but to accept the long-term direction of U.S. technological innovation while maintaining diversification and risk control.

If the AI revolution is real, the center of U.S. and global capital markets may shift even more decisively toward the United States.

At the same time, rates, exchange rates, valuations, and debt risks will remain volatile.

In the high-rate new normal, the real challenge is not whether to believe the AI growth story, but at what price and with what risk that story is owned.

< Summary >

ARK’s Cathie Wood interprets current high rates not as a risk signal but as evidence of AI-driven productivity growth.

She argues that the U.S. 10-year yield around 5% is not historically abnormal and that stronger growth could justify even higher yields.

The main innovation pillars are artificial intelligence, robotics, energy storage, biotechnology, and blockchain.

Falling AI usage costs may support both productivity gains and market expansion.

The strong-dollar thesis is based on global AI subscription spending and capital flows into U.S. companies.

However, long-duration bonds, highly leveraged growth stocks, stretched AI valuations, and strong-dollar risks remain important concerns.

For most investors, a diversified accumulation strategy focused on the Nasdaq 100 or broader U.S. equities may be more practical than selecting individual innovation stocks.

[Related Articles…]

• AI Infrastructure Spending Cycle and U.S. Big Tech Growth Outlook
• Rising U.S. Treasury Yields and Global Asset Market Adjustments

*Source: [ 소수몽키 ]

– 초고금리 뉴노멀 시대에 대비하라? 월가 초고수의 깜짝 예언, 적중할까

---

● AI-Memory Supercycle, Nasdaq Defies Yield Shock

U.S. Treasury Yield Breaks 5.3% While the Nasdaq Holds Up | Micron Earnings Call Signals an AI Memory Supercycle

The key message from this New York briefing is clear.

Even as the U.S. 10-year Treasury yield surged to 5.34%, the Nasdaq did not break down, with mega-cap tech and AI semiconductors providing support.

Micron’s earnings call delivered a strong signal that memory supply could remain tight through 2028, suggesting that this is not merely a cyclical semiconductor upswing but a structural shift in the memory industry driven by AI infrastructure investment.

At the same time, concerns over French sovereign debt, a 17-month low in the euro, dollar strength, higher oil prices, and a sharp decline in the probability of a Fed rate hike all point to a global financial market that appears resilient on the surface but remains highly sensitive beneath it.

Today’s note connects U.S. equities, Treasury yields, Fed policy, dollar strength, and the AI semiconductor cycle.

1. Market Overview | A Strange Environment of High Rates, High Oil, and a Strong Dollar, Yet Tech Holds Firm

U.S. equities are, on the surface, holding up reasonably well.

S&P 500 futures, Nasdaq futures, and Dow futures all opened in positive territory, with the Nasdaq showing relative strength.

However, the internal market picture is far from calm.

The U.S. 10-year Treasury yield rose as high as 5.34%, its highest level since 2002.

At this level, growth stocks and technology shares would normally be under significant pressure.

Higher discount rates reduce the present value of future earnings, creating direct valuation pressure on Nasdaq-style growth stocks.

This time, however, the Nasdaq has been supporting the market instead.

The reason is straightforward.

Expectations for earnings from AI-related mega-cap technology and semiconductor companies are temporarily outweighing the burden of higher rates.

• U.S. 10-year Treasury yield: as high as 5.34%
• WTI crude: around $90 per barrel
• Brent crude: back near $100
• Dollar index: continuing to strengthen
• Euro: at a 17-month low versus the dollar
• Nasdaq: relatively strong, led by mega-cap tech and AI names

In short, the market is being held together by AI-related optimism despite the combined pressure of higher rates, higher oil, and a stronger dollar.

2. U.S. 10-Year Yield Breaks 5.3% | Why Yields Rose Despite Softer Inflation

The most important variable in this market is clearly the U.S. Treasury yield.

The prior day’s PCE inflation data came in softer than expected.

Normally, easing inflation would reduce the probability of additional Fed tightening and push Treasury yields lower.

This time, that relationship did not hold.

Inflation data were benign, yet long-term Treasury yields moved sharply higher.

The reason is that the yield increase is not being driven solely by the Fed.

• Higher oil prices due to Middle East risk
• Rising U.S. government debt and increased Treasury issuance
• A U.S. economy that remains stronger than expected
• Weakening investor demand for long-duration Treasuries
• Broad selling across global bond markets

The supply side of the Treasury market is particularly important.

As the U.S. government increases issuance to finance large fiscal deficits, investors demand higher yields.

In other words, even without further Fed hikes, market rates themselves are creating tightening conditions.

Wall Street is now viewing 5.5% as the next major psychological resistance level for the 10-year yield.

BNP Paribas has argued that if the 10-year yield approaches 5.5%, pressure on equities could intensify.

With the yield already at 5.3%, the market is likely to remain highly sensitive to the 5.5% level.

3. Why Is the Nasdaq Rising at 5.3% Yields? | The Answer Is AI Earnings and Big Tech Balance Sheets

The key question is why the Nasdaq remains firm even as rates rise to this extent.

There are three main reasons.

3-1. Micron Confirmed Strong AI Memory Demand

Micron repeatedly emphasized very strong memory demand in its earnings call.

In particular, demand remains robust for HBM, high-performance DRAM, and SSDs used in AI servers.

This is not simply a strong quarter for one company.

It is a sign that AI infrastructure investment is translating into revenue.

Although Micron’s share price fell on short-term profit taking after earnings, the results helped support sentiment across semiconductors and AI-related equities.

3-2. Mega-Cap Tech Can Absorb Higher Rates

Companies such as Nvidia, Microsoft, Alphabet, Amazon, and Meta are relatively well positioned to withstand higher rates.

They generate strong cash flow and carry comparatively low debt burdens.

As a result, higher rates are less likely to threaten their financial stability.

The situation is different for unprofitable small- and mid-cap companies.

This also helps explain the limited performance of the Russell 2000.

In a high-rate environment, companies that already generate profits have a clear advantage over those that still depend on borrowing to grow.

3-3. The Market Is Asking Whether Earnings Growth Can Outrun Rates

Investors are no longer focused on rates alone.

They are assessing whether earnings growth from AI companies can offset the current level of rates.

In other words, if AI-related profits rise faster than the burden created by 5.3% yields, equities can remain supported.

However, if AI guidance weakens, current valuations could quickly become a liability.

4. A Key Point Often Missed Elsewhere | AI Companies Lift Equities While Also Pushing Yields Higher

The most important feature of this market is that AI companies are supporting equity prices while also contributing to higher long-term yields.

This is not widely emphasized in many reports.

AI companies are spending heavily on data centers, GPU clusters, power infrastructure, and network equipment.

That spending often requires bond issuance or large-scale financing.

The issue is that this capital demand competes with Treasury issuance.

Investor capital is finite, while both the U.S. government and AI infrastructure companies are drawing heavily from the same market.

As a result, bond investors demand higher yields.

AI therefore has a dual effect: it supports equities through growth expectations, but it can also place upward pressure on bond yields.

• AI earnings expectations → supportive for the Nasdaq
• Expanding AI data center investment → higher funding demand
• More corporate bond issuance → competition with Treasuries
• Higher yields → valuation pressure on equities

The market’s next direction will depend on whether AI earnings can continue to offset the pressure from rising yields.

5. October Fed Hike Probability Falls to 36% | Market Rates Are Already Acting as Tightening

The probability of a Fed rate hike in October has fallen sharply.

Just days ago, the probability was close to 70%; it is now around 36%.

The probability of a hold has risen to roughly 64%.

The main reason is the sharp rise in long-term yields.

With 10-year and 30-year Treasury yields at their highest levels since 2002, market rates are already tightening financial conditions.

Mortgage costs, corporate borrowing costs, and consumer credit costs are all rising.

That means the Fed may not need to raise rates further to achieve a tightening effect.

New York Fed President John Williams also signaled that there is no urgent need to accelerate rate hikes.

Following that statement, the market quickly reduced the implied probability of a hike.

The Fed is not in a fully comfortable position, however.

Oil prices remain elevated, the dollar is strong, and the labor market has not clearly softened.

The unemployment rate remains near full employment at around 4.1%, although there are also signs of slower hiring and weaker job-seeking conditions.

The Fed is likely to assess how much of a cooling effect higher market rates are already having before considering any additional move in October.

6. Why France Has Become a Central Bond-Market Risk | Concerns Over European Fiscal Stress

An unexpected source of stress in the global bond market is France.

The French government has proposed 43 billion euros in spending cuts and cost reductions next year.

The plan is intended to reduce the fiscal deficit from 5.4% of GDP this year to 5.0% next year.

The problem is that the market does not appear confident that the plan will be implemented.

French politics remain fragmented, and it is unclear whether austerity measures can pass parliament and be executed effectively.

This skepticism is pushing French bond yields higher.

• French 10-year yield: near 5%
• Highest level since 2002
• Widening spread versus German Bunds
• Planned French bond issuance next year: about $380 billion
• Interest expense: about 91 billion euros, up 15% from the previous year

France is a core euro area economy.

If French bonds weaken further, European yields could rise more broadly, increasing funding costs for European companies.

In that sense, French fiscal concerns are not just a domestic issue; they are a potential source of volatility for the wider European financial system.

7. Euro at a 17-Month Low, Dollar Strength Continues | Higher Oil Is Hitting Europe Harder

The euro fell below $1.13, reaching its weakest level in 17 months.

For September, the euro has declined by around 2.5%.

The main drivers are higher oil prices and pressure on the European economy.

Europe is highly dependent on energy imports.

When oil rises, both inflation pressure and growth pressure increase.

French fiscal concerns have further weakened sentiment toward the euro.

By contrast, the U.S. has remained supported by higher Treasury yields, which continue to attract capital.

Even though expectations for additional Fed hikes have eased somewhat, the dollar remains firm because U.S. assets still offer attractive yields.

In short, Europe is facing a combination of high inflation and weak growth, while the U.S. is facing high rates but still demonstrating economic resilience.

That divergence is reflected in euro weakness and dollar strength.

8. Micron Results | AI Memory Demand Was Confirmed by the Numbers

Micron’s results were one of the most important events in the market.

According to the release, Micron reported fourth-quarter revenue of $54.2 billion, up 31% sequentially and 379% year over year.

Adjusted EPS rose 33% sequentially, and gross margin reached 87%.

The company also noted six consecutive quarters of record revenue.

The next-quarter revenue guidance was $61.5 billion, with a strong EPS outlook.

What mattered most was not only the numbers, but the message from management.

Micron said memory supply should remain tighter in 2027 and 2028.

This suggests that the current memory upcycle may be more durable than a typical short-lived cycle.

9. Micron Earnings Call Key Point 1 | Supply Shortage Could Extend Through 2028

Micron’s CEO said the company expects demand to exceed supply in both 2027 and 2028.

He indicated that supply-demand conditions may become even tighter in 2027 and 2028 than in 2026.

This matters because it differs from the historical logic of the memory industry.

Traditionally, when prices rose, companies expanded supply, and over time that led to oversupply and price declines.

Micron argues that this cycle is different.

The reason is that HBM and AI memory demand are strong, while supply is difficult to increase quickly.

• New cleanroom construction takes time
• Higher HBM output reduces room for standard DRAM production
• AI server demand is rising faster than expected
• Customers are actively securing long-term volumes

Micron stated that there is no clear visibility on when supply will catch up with demand.

That point is important for how the market values the company.

Investors will need to decide whether Micron remains a cyclical peak-earnings name or whether it is becoming a structurally stronger AI infrastructure business.

10. Micron Earnings Call Key Point 2 | More Than 75% of 2027 Output Is Already Committed

Micron said that more than 75% of its 2027 production is already committed to customers.

That is a significant figure.

For a semiconductor company to have already secured this much volume more than a year ahead suggests that customers are taking the supply shortage very seriously.

Even more importantly, discussions are already moving toward 2028 volumes.

This indicates not just interest, but actual demand for additional supply and orders.

If this structure continues, Micron’s earnings volatility may decline compared with the past.

Historically, memory-company results swung sharply with price cycles, but greater use of long-term contracts and advance orders would improve earnings visibility.

11. Micron Earnings Call Key Point 3 | HBM Price Increases and Margin Expansion

HBM is the most important memory product in the AI semiconductor market.

Micron said that most of its 2027 HBM volume is already sold and that prices are higher than in 2026.

HBM has long been seen as a major revenue driver, but not necessarily as a major profit contributor.

That changes if pricing continues to improve.

Micron said the margin gap between HBM and standard DRAM is narrowing.

In other words, HBM could increasingly support not only revenue growth but also margin expansion.

This has important implications not only for Micron, but also for other memory companies such as SK hynix and Samsung Electronics.

12. Micron Earnings Call Key Point 4 | Lower HBM Specifications Do Not Necessarily Signal Weak Demand

There have been reports that some major customers are lowering HBM specifications or adjusting memory content.

On the surface, that could suggest that HBM demand is weakening because it has become too expensive.

Micron’s CEO, however, said the company does not view this as a demand deterioration.

According to management, customers are adjusting specifications not because they need less memory, but because they want to build more systems with limited memory supply.

For example, reducing memory per server allows more servers to be built from the same amount of memory.

This is better understood as a supply constraint response than as a sign of weakening demand.

AI models continue to grow, context windows are getting longer, and workloads are expanding.

Micron therefore believes underlying memory demand remains strong.

13. Micron Earnings Call Key Point 5 | Long-Term Supply Agreements Are Changing the Memory Cycle

Micron said it has signed 26 long-term supply agreements, and that more than 35% of revenue through 2030 is already covered by such contracts.

The CEO said this share could eventually rise to around 50%.

This is a major shift.

The traditional memory cycle was so volatile because future demand was difficult to forecast.

Companies would build capacity based on strong demand, only to face weaker demand 1–2 years later, leading to oversupply.

Inventories would rise, prices would fall sharply, and both earnings and share prices would weaken.

With more long-term supply contracts, companies can align capacity planning more closely with customer demand.

This represents a structural change in the memory industry.

That is why management described the shift as a fundamental change in the industry.

14. Joint Development with Nvidia on NV-HBM | From Supplier to Design Partner

Micron said it has been co-developing NV-HBM with Nvidia for more than a year.

This is important.

Traditionally, memory companies produced standard products and supplied them to customers.

NV-HBM, however, is closer to a custom HBM solution developed together with Nvidia from the design stage.

In that structure, customers are less likely to switch suppliers easily because the specifications are co-engineered.

For Micron, this improves product differentiation and pricing power.

For Nvidia, it secures optimized memory supply.

Over time, the relationship between Micron and Nvidia may evolve from a conventional supplier-customer model into a strategic partnership.

15. Technology Gap Versus Chinese Memory Producers | Micron Says It Is at Least Two Nodes Ahead

Competition from Chinese memory producers remains an important issue.

China continues to invest heavily in memory semiconductor self-sufficiency.

Micron said, however, that it is at least two process nodes ahead of Chinese competitors in advanced DRAM.

The company said its 1-gamma DRAM already accounts for more than half of bit output, and that development of the next-generation 1-delta node is progressing well.

Micron also emphasized that experience with EUV in advanced DRAM is becoming increasingly important, making it difficult to close the gap quickly.

The company added that its China revenue exposure is declining and could fall to low single digits by around 2027.

In other words, China risk remains, but the technology barrier in high-value AI memory appears still to be substantial.

16. Rising CAPEX and the Oversupply Risk | Micron’s Explanation Centers on Cleanrooms, Not Equipment

One of investors’ main concerns is rising capital expenditure.

When memory companies begin generating strong profits, they often overinvest, eventually creating oversupply a few years later.

Micron said investors should distinguish between construction CAPEX and equipment CAPEX.

A large part of the current increase is tied to cleanroom construction rather than production equipment.

Building cleanrooms does not immediately add output.

Micron said the decision to add tools and begin wafer production will depend on demand at that time.

In other words, the company is securing options for future capacity rather than immediately flooding the market with supply.

If this explanation is convincing, oversupply concerns could ease somewhat.

17. Agentic AI Is Expanding Demand for DRAM and SSDs

AI demand is not limited to HBM.

Micron said the spread of agentic AI could increase demand for CPUs, DRAM, and SSDs as well.

Agentic AI refers to systems that not only answer prompts but also understand user objectives and perform multi-step tasks independently.

For these systems to operate at scale in enterprise and consumer environments, they require servers, storage, memory, and networking infrastructure.

Micron said these trends are contributing to high-teens growth in server shipments.

In that sense, the AI memory cycle is broader than HBM alone.

It should be viewed as a wide-ranging demand cycle covering HBM, high-performance DRAM, NAND, SSDs, and data-center storage.

18. Investment Takeaways | Five Points to Watch

From an investment perspective, there are five key points to monitor.

18-1. Whether the U.S. 10-Year Yield Breaks 5.5%

The market is currently absorbing 5.3%, but 5.5% would be a different matter.

If the 10-year yield breaks above 5.5%, valuation pressure across equities could intensify.

18-2. AI Mega-Cap Earnings Guidance

If AI companies continue to deliver strong earnings, the market can withstand higher rates.

But if monetization of AI investment by Nvidia, Microsoft, Alphabet, Meta, and Amazon slows, a deeper correction is possible.

18-3. Whether Micron’s 2027–2028 Supply Shortage Outlook Holds

Micron’s long-term supply shortage outlook remains a key question.

If supply increases faster than expected, or demand slows, memory stocks could face pressure.

18-4. French Bond Yields and European Spreads

If France’s fiscal stress spreads across Europe, global bond-market volatility could rise further.

That could support the dollar and weigh on risk assets.

18-5. Oil Prices and the Risk of Inflation Re-Acceleration

If oil remains elevated, expectations for disinflation could weaken.

That would allow market yields to rise again even if the Fed keeps rates unchanged.

19. Conclusion | AI Is Supporting Markets, While Bonds Are Driving Volatility

Current U.S. equities are not simply in a standard bull market.

AI semiconductors and mega-cap technology are supporting the major indices, while U.S. Treasury yields and global bond-market stress continue to drive volatility.

Micron’s earnings call suggests that AI memory demand may be part of a long-term infrastructure cycle rather than a short-lived trend.

In particular, more than 75% of 2027 production already committed, the possibility of shortages through 2028, co-development of NV-HBM with Nvidia, and expanding long-term supply agreements point to a structural shift in the memory industry.

However, the risks remain clear.

If the 10-year yield moves toward 5.5%, even strong AI earnings may not be enough to offset the broader market impact.

French sovereign risk, euro weakness, higher oil prices, and dollar strength also need to be monitored closely.

At this stage, the key question is not whether AI can overcome all negative factors, but how far AI earnings can offset the pressure from higher rates.

• Treasury
• Micron

*Source: [ Maeil Business Newspaper ]

– 금리 5.3%인데 증시는 왜 오를까｜마이크론 어닝콜 완전분석｜빅테크 AI 메모리 선점｜연준 인상 확률 36%로 뚝｜채권 진앙지 된 프랑스｜유로 최저·달러 강세｜홍혜진의 뉴욕브리핑

---