My kids love playing four square at recess, and I remember loving it too when I was in elementary school. So naturally, we started playing at home in the driveway. It’s a lot of fun. But it can also get competitive, and it usually ends with someone crying (and no, it’s not me). If you’ve never played, the rules are simple. There are four squares, labeled A through D (or 1-4). The ball bounces into your square, and you hit it into someone else’s. Miss it, and you’re out; everyone below you moves up a spot. The goal is to get to the A square and stay there as long as you can. Why? Because A serves and everyone else spends the game reacting to whatever comes their way.
But we’re not going to spend the entire update talking about our heated driveway competitions. This week, four of the biggest names in technology took their turns on the court. Microsoft (MSFT), Alphabet (GOOG), Amazon (AMZN), and Intel (INTC) all reported earnings, and on paper most of the numbers were strong. Revenue was up across the board, and cloud businesses are growing at rates we haven’t seen in years. And yet three of the four stocks got hammered, with Alphabet posting its worst day ever as a public company and Intel its worst in over 50 years.
So what happened? The market may be moving from asking whether AI is real to asking who can afford it. Microsoft, Alphabet, and Amazon are on pace to spend nearly $600 billion between them on AI infrastructure this year. Two of them are now spending more than their businesses actually bring in, and Alphabet is already covering the gap with borrowed money and new stock. Even Intel, which isn’t building data centers, got caught in the crossfire. So let’s take them one at a time (the numbers, the guidance, and how the stock reacted) and rank them the only way that feels appropriate this week: by square.
A Square: Microsoft
Microsoft held its spot this week, and it’s a useful baseline for everything that follows. Revenue came in at $92.3 billion, up 19%, and adjusted EPS of $4.89 beat the $4.73 estimate. The highlight was Azure (their cloud business), which grew 48% to $33.2 billion. This was the first time Microsoft has disclosed Azure revenue in actual dollars rather than just a growth rate. Finally, commercial remaining performance obligation (essentially, signed contracts not yet delivered) jumped 73% to $677 billion.
Now, on to the spending, which was enormous. But it was covered. Capital expenditures and finance leases totaled $45.2 billion, up about 30% from a year ago, while free cash flow fell 27% to $18.6 billion. Down, yes. But still positive, which this week turned out to be the report’s most important number.
Guidance was strong as well. Microsoft projected $96 billion to $97.1 billion in fiscal second-quarter revenue and 49% to 50% Azure growth, with CFO Amy Hood noting that demand still exceeds supply. Hood also said the company expects to stay free cash flow positive even as capex climbs toward roughly $170 billion.
Shares rose 4% in extended trading, then gave it back after the company disclosed a bond offering of at least $25 billion, its first major U.S. bond sale since 2017. The stock closed Thursday effectively flat. In a week like this, though, flat is a win. But it’s worth noting that even the kid in A square is now borrowing lunch money.
B Square: Amazon
Amazon’s operations were outstanding. AWS grew 39% to $46 billion, its fastest pace in nearly five years, and operating income of $28.1 billion produced a 13.9% margin, the highest in company history. The AWS backlog reached $578 billion in contracted future revenue, more than double last year.
But diluted EPS of $1.89 missed the $1.96 consensus, and the cash flow statement told the real story. In Q3 alone, operating cash flow was $42.4 billion against $75.9 billion in net cash capex, a quarterly free cash outflow of roughly $33.5 billion. On a trailing twelve-month basis, free cash flow swung to an outflow of $42.4 billion from an inflow of $14.8 billion a year earlier.
And there’s more spending coming. Management still expects roughly $220 billion of cash capex in 2026 and plans to spend meaningfully more in 2027. Q4 guidance called for revenue of $240 billion to $247 billion and operating income of $28.5 billion to $33.5 billion. The stock fell about 7% in premarket trading Friday; however, it has pared that to a couple of percent as of this writing. That’s painful, but compared to its neighbor in C, Amazon got off relatively easy.
C Square: Alphabet
Alphabet’s business appears to be in great shape. Revenue rose 25% to $127.6 billion, and Google Cloud more than doubled to $30.7 billion, up 103%.
However, the bottom line went the other way. EPS of $2.12 came in well short of the $3.02 consensus, and down from $2.87 a year ago. This was driven largely by an $8.7 billion paper loss on equity investments that cut 55 cents per share. That part is mostly noise. What’s harder to dismiss is Search, which grew only 12%, down from 17% in Q2 and 19% in Q1. And Search is still the engine that funds everything else at Google. So not only was growth slower, it fed into the narrative that AI may be impacting their cash cow more than previously believed.
And about that cash. Alphabet generated $42.3 billion in operating cash flow, but capex totaled $51.4 billion, leaving a $9.1 billion free cash flow deficit. To fill the gap, Alphabet raised about $12.4 billion through stock issuance and $28.7 billion in debt during the quarter. And it isn’t slowing down: capex is expected to reach about $200 billion in 2026 and rise to $300 billion in 2027.
Investors didn’t love that combination. Shares closed down about 13% on Thursday, the largest one-day drop in the company’s history (it went public in 2004), erasing roughly $500 billion in market value. That’s more than the entire market cap of most companies in the S&P 500, gone in a single session.
D Square: Intel
Intel is a different story, but it connects to the same theme. The quarter itself was good: adjusted EPS of 52 cents beat the 41 cents expected, on revenue up 21% to $16.5 billion, and data center and AI revenue rose 78% to $7.3 billion.
The problem was the outlook. Intel guided fourth-quarter revenue of $15 billion to $16 billion versus the $17 billion analysts expected, and adjusted EPS of 25 cents versus 43 cents, with gross margin shrinking from 43% to 38%. The two main culprits were a memory chip shortage hurting PC demand and an accidental factory shutdown in Israel. Intel now expects the PC market to shrink about 12% in 2026 and another 10% in 2027.
That memory shortage is the link back to the other three. When hyperscalers buy (almost) every memory chip they can get their hands on for data centers, fewer (and more expensive) chips are left for laptops. Intel didn’t miss the ball so much as get hit by one aimed at someone else. The stock fell almost 30% in trading Friday. It’s just another massive price move in a year that has seen the stock go from $40 to $140 to $65. Hold on to your hats.
Who Gets to Serve?
The common thread isn’t that AI is a bubble or that these companies are in trouble. Alphabet and Amazon just posted strong revenue with exploding backlogs. The thread is the shift in how the buildout is being funded and perceived. For years, the hyperscalers paid for their ambitions out of pocket. Now Alphabet and Amazon are running negative free cash flow.
That matters more than it might seem. A company funding growth from its own cash flow is serving: it sets the pace and answers only to its own results. A company funding growth by issuing bonds and stock is returning serves, and it answers to the capital markets. Microsoft is borrowing too, but there’s a difference between choosing to borrow and needing to. And capital markets have moods (just like my kids playing four square). They’re generous when things are going well and much less so when they’re not. This week’s split verdict (Microsoft spared, Alphabet and Amazon punished) wasn’t really about who has the better AI. It was about who can still pay for it without asking for help.
Back to the Driveway
None of this means the AI buildout will fail. The demand is real, the contracts are signed, and the revenue is showing up. But “transformative technology” and “great investment at any price” are two very different things, and even the best companies can’t spend more than they earn forever without the market asking them to show their work.
The thing about four square is that nobody stays in A forever, and nobody stays in D forever either. The squares rotate constantly, sometimes several times in a single recess. The same is true of market leadership. For long-term investors, the takeaway from this week isn’t to chase whoever’s in A or abandon whoever just got bounced. It’s to stay diversified, keep position sizes sensible, and not let one rough week (or one great one) dictate your plan. Tears optional.
Markets / Economy
- Things came to a head Thursday and Friday as big-tech earnings weren’t what the market was looking for. The S&P finished the week down -0.6%, the Nasdaq down -2.0%, and the small-cap Russell 2000 up 1.3%.
- The U.S. economy expanded an annualized 3.4% in Q3 2026, the most in a year. This compared to a 2.2% rise in Q2 and expectations of 3.2%. Private domestic investment drove the largest upward contribution, supported by strong spending in the AI buildout.
- The U.S. PCE price index rose 0.5% MoM in September, the most in six months, following a 0.3% increase in August. However, while elevated, it was in line with market expectations.
Stocks
- U.S. equities were in negative territory. Communication Services and Technology led the decline, while Healthcare and Materials outperformed. Value stocks led growth stocks, and small caps beat large caps.
- International equities closed higher for the week. Developed markets fared better than emerging markets.
Bonds
- The 10-year Treasury bond yield decreased six basis points to 4.94% during the week.
- Global bond markets were in positive territory this week.
- Government bonds led for the week, followed by high-yield bonds and corporate bonds.
