TechFlow Logo
Login/ Sign up
ETH Gas
Gwei
Fear
gas
Strong Earnings, Stock Price Falls: Who Is Footing the Bill for AI Infrastructure?

Strong Earnings, Stock Price Falls: Who Is Footing the Bill for AI Infrastructure?

2026.07.30
Share

TechFlow Selected TechFlow Selected

techFlow

Strong Earnings, Stock Price Falls: Who Is Footing the Bill for AI Infrastructure?

Is it the company's own cash flow, customer prepayments, or debt?

2026.07.30 - 05:39:57
AI
Is it the company's own cash flow, customer prepayments, or debt?

Author: Alea Research

Translation: TechFlow

TechFlow Editor's Note: Last week, almost all earnings reports confirmed that AI computing power demand is real, yet large-cap tech stocks fell instead. The market is repricing based on a new standard: Who will fill the funding gap for AI infrastructure—is it the company's own cash flow, customer prepayments, or debt? This determines which companies can last until the day AI monetizes.

Growing Companies Were Sold Off, Where Is the Funding Gap, and Who Pays for AI Infrastructure

Last week, most companies reporting earnings saw revenue growth, yet stock prices still fell. The S&P 500 fell 0.6%, and the Nasdaq fell 2.1%, almost every earnings report confirmed that AI computing power demand is real.

Building AI capacity means paying for chips, data centers, memory, and power years in advance, with revenue arriving much later. The market spent a week categorizing companies: looking at who fills this gap—the company's own cash, customer money, or borrowed money.

This article is the equity section of this week's Pulse report.

image

Figure: Performance of Asset Classes Over the Week: Energy Leads, Long-Duration Assets Under Pressure

Source: Alea Research / Substack

Why Stocks Fall Even on Good News

SPY is the market-cap weighted S&P 500, dominated by a few giant tech stocks, up 0.1% for the week. RSP holds the same 500 stocks but with equal weight for each, up 0.8%. QQQ is the tech-concentrated Nasdaq 100, down 1.1%, and the chip stock basket SOXX fell 4.3%. The broader market was bought on Friday; the sell-off was specifically targeted at large AI concept stocks.

image

Figure: AI Complex Return Distribution: Semiconductors One-Year Return 129%, Far Higher Than Nasdaq 100's 22%

Source: Alea Research / Substack

When interest rates look harder to cut, these stocks fall the hardest because their prices depend on expected profits years down the line. When rates rise, holding boring alternatives like bonds earns more just by waiting, so every future profit must be discounted against this richer alternative. Investors are not exiting the market—if they were, the equal-weight basket would also fall. They are shedding stocks whose value sits furthest in the future, just as the future has become expensive.

The Funding Gap

A company generates cash from operations and also spends cash on capital expenditures—chips, buildings, and grid access needed for infrastructure. When capital expenditures exceed operating cash flow, the difference must come from somewhere, either new debt, new equity, or the company's reserves.

All three became more expensive last week. The cost of debt is directly the interest rate. The cost of equity financing is higher because interest rates push stock prices down, meaning companies must sell more shares to raise the same amount of money. Spending reserves means forfeiting the interest these reserves could earn now, and interest rises with rates.

Interest rates face two major upward pressures. First, attacks on oil tankers disrupted transport on key routes, pushing Brent crude over $100 per barrel, raising baseline inflation expectations. Second, import prices rose 7.1% year-over-year, driven by cost increases in computers, semiconductors, and industrial machinery. This is exactly the hardware driving AI infrastructure expansion. These pressures limit the Federal Reserve's room to ease monetary policy, with interest rate futures pricing showing a 75% probability of staying unchanged, 25% probability of a rate hike.

image

Figure: Technical Pattern of Alphabet After a 10% Drop in a Single Week

Source: Alea Research / Substack

Alphabet was strong across all businesses this quarter: Cloud revenue grew 82% to $24.8 billion, Cloud margin expanded from 20.7% to 35.6%, and Search still grew 17%. The stock still fell 7.2% after earnings because Alphabet spent $44.9 billion on capital expenditures, while operating cash flow was only $39.1 billion, a single-quarter gap of $5.9 billion, and then told investors it expects 2026 capital expenditures to be $195-205 billion. No one questions demand, so the sell-off is repricing how the spending is financed.

image

Figure: Alphabet Compared to Microsoft with Stronger Cash Conversion Capability

Source: Alea Research / Substack

Oracle reported an annual funding deficit, $55.7 billion in capital expenditures versus $32 billion in operating cash flow (a gap of $23.7 billion), with a financing plan requiring a combination of debt and equity issuance of approximately $40 billion. Despite securing a Pentagon contract worth up to $7 billion, the stock price still touched a 52-week low because money must be spent building data centers before revenue arrives.

image

Figure: Oracle Stock Price Touches 52-Week Low While Capital Expenditure Plan Continues to Expand

Source: Alea Research / Substack

Tesla delivered a record 480,126 vehicles, but operating margin compressed to 1.4%. Operating cash was insufficient to cover $5.8 billion in quarterly infrastructure spending, resulting in negative free cash flow of $1.1 billion.

image

Figure: Tesla Q2 Sales Rebound, But Final Free Cash Flow Negative

Source: Alea Research / Substack

Companies Where Customers Pay in Advance

Companies that preserved market cap all have customers or counterparties who committed via contract to absorb part of the infrastructure costs. First, customer deposits are cash handed over now for future supply, directly financing factories with buyer money. Second, take-or-pay agreements require customers to pay an agreed minimum amount regardless of whether they take delivery, turning future demand into a legal right lenders are willing to lend against. Finally, backlog, which is the pile of signed but undelivered orders, does not provide funding itself, but it eliminates the demand risk that would cause lenders to charge higher fees.

Micron has 16 multi-year take-or-pay agreements, covering about one-fifth of DRAM output until 2030, approximately $100 billion in contractual minimum revenue, and $22 billion in customer deposits. Its customers are paying for it to build factories.

image

Figure: Absolute Values of GE Vernova Relative to Industrial Peers

Source: Alea Research / Substack

GE Vernova sells turbines and grid equipment, converting electricity demand into power, booking $24.2 billion in orders in a quarter that delivered $11.1 billion in revenue; its $176 billion backlog equals about four years of sold work.

Lockheed's backlog reached $230 billion, about 2.9 years of revenue. Microsoft is the self-funded version of the same escape route: it spent $31.9 billion on capital expenditures, approximately 68% of $46.7 billion in operating cash flow, and still had $15.8 billion in cash remaining after infrastructure. It is running the same infrastructure as Alphabet but does not need external funding.

image

Figure: Book-to-Bill Comparison of Lockheed and Northrop

Source: Alea Research / Substack

There are two points to note. Pre-sold revenue is not pre-sold profit. Northrop booked $1.84 in new orders for every $1 delivered, yet profit margin still slipped from 11.8% to 10.6%, because orders may arrive faster than the company can profitably deliver. Moreover, diversification among these companies is thinner than the list of stock tickers suggests: most incremental order growth in memory, chips, and power equipment traces back to the budgets of the same few data centers. This is why Micron, despite having contracts, still fell 6.9% on Friday. Holding memory manufacturers, chip baskets, and turbine manufacturers adds stock tickers faster than it adds diversification.

Microsoft and Amazon Earnings

Microsoft reports earnings on July 29; only if Azure growth is close to 40%, operating income grows faster than revenue again, and capital expenditures climb above $40 billion per quarter, can it remain in the self-funded category. The Federal Reserve also announces interest rate decisions on July 29, so volatility is expected.

image

Figure: Microsoft Underwent a Year of Valuation De-rating Before July 29 Earnings

Source: Alea Research / Substack

Amazon reports earnings on July 30; tracked free cash flow has dropped from $25.9 billion a year ago to $1.2 billion. Its own guidance implies an 11.2% operating margin, versus 13.1% last quarter; if operating income exceeds the $22 billion figure, it will show that the cloud business is filling cash faster than infrastructure consumes it.

image

Figure: Amazon's Path Before July 30 Earnings

Source: Alea Research / Substack

Join TechFlow official community to stay tuned

Add to Favorites
Share to Social Media

Related Articles

2026.07.29

Mining Giant, Raises 2.7 Billion: Switches to AI Landlord, IPOs

Ionic Digital (IOND) Transforms from Celsius Bankrupt Mining Farm into AI Data Center Operator, Surges 26% on NASDAQ Debut on July 28

Mining Giant, Raises 2.7 Billion: Switches to AI Landlord, IPOs
2026.07.29

Morgan Stanley Research Report Analysis: AI Infrastructure ROIC Can Reach Up to 40%, Companies Building Their Own Compute Are the Biggest Winners

If the 40% ROIC is true, then these hundreds of billions of dollars in capital expenditure are creating value.

Morgan Stanley Research Report Analysis: AI Infrastructure ROIC Can Reach Up to 40%, Companies Building Their Own Compute Are the Biggest Winners
2026.07.29

Americans' Disappointment with AI: Hype Bubble Bursts, Middle Class Is Paying for Super Data Centers

More and more people are realizing that this trillion-dollar gamble is harming the interests of ordinary people, while Silicon Valley giants and Wall Street elites are frantically cashing out.

Americans' Disappointment with AI: Hype Bubble Bursts, Middle Class Is Paying for Super Data Centers
2026.07.29

Podcast Notes | OpenAI CEO Altman's Latest Interview: I Am Not At All Anxious About Open-Source AI Competition, Very Confident in the Upcoming Model

We are about to create a magic lamp that can grant any wish. But the concentration of power brought by AI is a terrifying thing.

Podcast Notes | OpenAI CEO Altman's Latest Interview: I Am Not At All Anxious About Open-Source AI Competition, Very Confident in the Upcoming Model
2026.07.29

The person who killed the middleman is now acting as a middleman for AI

Multiple crypto enterprises are leveraging their advantages in power and data center infrastructure to pivot to AI. Capital is flowing to AI, and the industry is shifting from disintermediation to becoming AI intermediaries.

The person who killed the middleman is now acting as a middleman for AI
2026.07.28

OKX.AI In-Depth Test: High Sales and High Ratings, Can It Withstand Scrutiny?

Does this market truly exist? Does Agent payment actually have value?

OKX.AI In-Depth Test: High Sales and High Ratings, Can It Withstand Scrutiny?
2026.07.28

IOSG | From Hot Storage to Cold Memory: Decentralized Storage Amidst the Storage Boom in the AI Era

AI storage is being frantically revalued by capital amidst the current wave of tech narratives.

IOSG | From Hot Storage to Cold Memory: Decentralized Storage Amidst the Storage Boom in the AI Era
2026.07.28

US Semiconductor Stocks "Black Monday": CDS Surge Exposes AI Financing Anxiety

The essence of this round of decline is the market's first collective scrutiny of the sustainability of the AI financing model.

US Semiconductor Stocks "Black Monday": CDS Surge Exposes AI Financing Anxiety
2026.07.28

Nvidia Guarantees $750 Billion for AI, Default Insurance Costs Hit Record

What is more worth watching than the stock price is the reaction of the credit market.

Nvidia Guarantees $750 Billion for AI, Default Insurance Costs Hit Record
2026.07.28

Morgan Stanley Research Report Analysis: AI Computing Power Gap Continues to Widen, Five Main Themes Present Investment Window

Short-term stock price volatility is mostly sentiment noise, while the imbalance in computing power supply and demand is the main theme for the medium to long term.

Morgan Stanley Research Report Analysis: AI Computing Power Gap Continues to Widen, Five Main Themes Present Investment Window
TechFlow Logo

Navigating Web3 tides with focused insights

Contribute An Articleemail
Media Requestsmsg

Risk Disclosure: This website's content is not investment advice and offers no trading guidance or related services. Per regulations from the PBOC and other authorities, users must be aware of virtual currency risks. Contact us / [email protected] ICP License: 琼ICP备2022009338号