Why tonight's earnings matter far beyond two companies—and what they mean for every investor.
KEY TAKEAWAYS
Tonight's Alphabet and Tesla earnings represent a broader referendum on whether massive AI investments are generating meaningful financial returns. Investors are looking beyond earnings beats and focusing on capital allocation efficiency.
Alphabet's rapidly expanding AI infrastructure spending has become the central issue. Cloud growth remains impressive, but management must demonstrate that escalating capital expenditures will ultimately translate into profitable growth.
Tesla's record deliveries are overshadowed by uncertainty surrounding profitability. The market remains focused on automotive margins and whether the company's long-term autonomous driving narrative remains credible.
External pressures–including weakness in semiconductor stocks, China's increasingly competitive AI models, and concerns raised by a leaked Treasury analysis–have increased scrutiny on AI spending. Market confidence has become noticeably more fragile.
The outcome of tonight's reports will likely influence investor sentiment toward the broader AI ecosystem. Subsequent technology earnings will be evaluated through the same lens.
MY HOT TAKES
Markets have moved beyond rewarding ambitious AI spending alone. Investors increasingly want measurable returns on the enormous capital being deployed.
The AI investment cycle has entered a new phase where execution matters more than vision. Great stories alone are no longer sufficient to support premium valuations.
Capital expenditure itself is not the problem. The problem emerges when spending persistently outpaces evidence of sustainable monetization.
AI optimism remains intact, but the burden of proof has shifted decisively onto management teams. Future earnings calls will increasingly focus on return on invested capital rather than AI ambitions.
The biggest risk is not disappointing quarterly earnings–it is weakening confidence in the broader AI investment narrative that has driven markets for the past two years.
You can quote me: "Tonight isn't an earnings season announcement–it's an audit."
Got faith? Your retirement account has a big date tonight! After the closing bell today, Alphabet and Tesla both step up to the mic and report earnings–and for once, this isn't a routine "did they beat the number" exercise. It's the first real gut-check for the trillion-dollar question hanging over every tech portfolio–every S&P-like portfolio–in America: is all that AI spending actually working, or are we watching the most expensive experiment in corporate history unravel before our eyes?
Here's why tonight matters more than usual. Alphabet–which I affectionately still refer to as Google mostly–is walking into this announcement already bruised with shares down over 6% over the past week, largely because of a math problem that's hard to spin. The company's capital spending is growing faster than its revenue. Not slightly faster. Significantly faster! We're talking about CAPEX nearly doubling year over year while sales grow a very respectable, but not-doubling, 21%. When the spending line outruns the sales line, eventually somebody asks the obvious question: what exactly are we buying with all that money?
Tesla's got its own version of the same anxiety, dressed up differently. The company just posted record deliveries–480,000+ vehicles, obliterating estimates–which sounds like…well, unambiguous good news. But scratch the surface and Wall Street can't agree on what those record sales are actually worth. Analyst profit estimates range from the mid-20-cent area to the mid-70-cent area per share, a gap wide enough to drive a Cybertruck through, because nobody's sure whether all that volume came with real margin or just deeper discounts.
Put those two stories side by side and you get tonight's real headline: this isn't an earnings season–it's an audit. And it's landing at the exact moment the rest of the market is already a bit on edge–chip stocks just tumbled into bear-market territory, there's a leaked and disputed Treasury memo floating around warning about AI-sector risk, and China just dropped a cheap new AI model that's got everyone asking why we're spending trillions if someone else can do it for less. Tonight, two of the companies at the center of the AI trade have to show their work. Let's ponder what happens if the math doesn't check out.
So let's start with Alphabet, because the core problem here doesn't need cherry-picked math to make the point. The company has already told us, on its own earnings call back in April, that full-year 2026 capital spending could run as high as $190 billion–and that it expects to increase that again in 2027. Meanwhile, Wall Street's revenue growth estimates for the quarter sit in the mid-20% range. When your own spending guidance is climbing toward $190 billion while your top-line growth is holding in the twenties, the gap between those two lines is the whole story–you don't need to dress it up with decimal points to see it.
Cloud is the number bulls will point to first–growth estimates cluster well north of 50%, and some analysts peg it even higher–and that's genuinely impressive. But "impressive growth in one division" doesn't automatically answer the bigger question hanging over the stock: does all that infrastructure spending eventually pay for itself, or is Alphabet just building the world's most expensive data centers and hoping the demand shows up in time to justify the bill?
To be fair, this isn't reckless by accident. Cloud growth is real. AI infrastructure has a shelf life measured in years, not quarters. But when a company's own guidance keeps climbing–$190 billion and rising–you'd better have a very good answer for what happens if AI demand growth even just plateaus. Tonight, on the call, somebody's going to ask that question in a much politer way. Watch how Sundar answers it…or doesn't.
Alphabet also didn't exactly get a clean runway into tonight. On July 16, word broke that Gemini 3.5 Pro–the company's next flagship AI model, and a big part of the pitch for why Cloud keeps growing–is running behind schedule. Shares fell on the news. Stack that on top of the €4.1 billion ($4.7 billion) EU antitrust fine on the Android case that Europe's top court upheld on July 2, closing out an eight-year legal fight, and you've got a stock that walked into this print with more than just a spending problem.
Then there's Tesla, which is its own special flavor of…well, so confusing. On the surface, this should be an easy win. Record deliveries–480,126 vehicles, up 25% year over year, smashing the roughly 403,000 the Street was expecting. Record energy storage too, at 13.5 gigawatt-hours, a big jump from Q1. If you just read the delivery number, you'd think Tesla's having its best quarter in years.
But Wall Street's EPS estimates tell a totally different story. Analyst estimates span a surprisingly wide range–which tells you the Street isn't remotely aligned on whether those record deliveries came with real margin or just deeper discounts and cheaper financing to move the metal. Automotive gross margin, excluding regulatory credits, was already thin last quarter–the kind of number that makes "record deliveries" look a lot less triumphant once you follow it down to the bottom line. The energy business runs noticeably fatter margins, but it's still a small slice of total revenue. My regular followers know that I keep reminding you: it’s a car company–cars remain the business. And cars remain the question mark.
And then there's the valuation elephant in the room that never shows up on the delivery number: Tesla trades at a sky-high multiple of earnings that has nothing to do with selling cars. You don't get a multiple like that from moving vehicles–you get it from investors betting on robotaxis, on Cybercab, on a driverless future that's been "coming soon" for what feels like a decade now. Oh yeah, robots too. 🤖 Management has already tried to temper expectations that robotaxi revenue shows up in any real way this year. So tonight's call isn't really about the cars at all–it's about whether Elon gives Wall Street a fresh reason to keep believing in the story that isn't on the balance sheet yet.
Here's the thing though–neither of these companies is reporting in a vacuum tonight. They're stepping into a market that's already jumpy, and for good reason.
Now, about that Treasury story–let me give you the accurate version, because it's gotten twisted in the retelling all over social media this week. A draft analysis out of the Treasury Department, reportedly written by a staffer rather than a formal department position, compared the current AI investment boom to the dot-com bubble, warning that AI companies are now so woven into the broader economy–and carrying so much institutional money along for the ride–that a serious pullback could ripple through stock markets, private credit, data center financing, cloud providers, chip makers, and utilities. Here's the part that matters: Treasury itself disowned it almost immediately. A spokesperson said the analysis "does not reflect the Treasury Department's views" and called it unvetted. And notably, even the leaked draft reportedly concluded the fallout would be less severe than the dot-com crash, not worse. So take it for what it actually is–a disputed, worst-case internal memo that never became policy–not an official warning shot from Washington. Still, the fact that a document like this exists at all, while Secretary Bessent has spent the year publicly championing something like $750 billion in AI infrastructure spending, tells you the anxiety is real even inside the building that's supposed to be cheerleading this thing.
Then, right on cue, the chip sector took it on the chin. The semiconductor trade tumbled into bear-market territory in mid-July as a red-hot, AI-fueled rally came undone in a matter of days. What set it off? China's Moonshot AI dropped a new open-source model called Kimi K3, and the reaction felt like the most intense bout of AI panic since DeepSeek rattled Wall Street back in January 2025. Nvidia's Jensen Huang pushed back hard, arguing–like he did with DeepSeek–that cheaper, more efficient Chinese models will ultimately increase demand for chips and data centers rather than kill it. Maybe he's right. But the fact that the market keeps flinching every time China ships something new tells you how fragile the AI capex thesis feels right now, even to the people who are all-in on it.
That's the backdrop tonight. Not a quiet Wednesday. A market already asking whether the AI emperor has clothes, right as two of its flagship believers step up to answer for their spending.
So what actually happens after the bell? Scenario one: Alphabet shows Cloud growth accelerating fast enough to make that $190 billion capex number look forward-looking instead of reckless, and Tesla posts margins closer to the optimistic end of the analyst range, with Musk offering something concrete–real dates, real numbers–on robotaxi expansion. If that happens, tonight becomes the relief rally that steadies the whole AI trade heading into Intel's report later this week and the rest of Big Tech earnings.
Scenario two: either company shows spending accelerating while margins or growth disappoint, even slightly. In that world, tonight doesn't just hurt Alphabet and Tesla shareholders–it hands fresh ammunition to everyone already circling the AI bubble thesis, and it could drag the chip sector deeper into the bear market it's already sitting in.
Tonight’s bar isn't "beat the number." It's "prove the number means something." Alphabet can beat on Cloud revenue and still lose the narrative if CAPEX keeps outrunning sales. Tesla can smash delivery records and still spook the market if the margin story doesn't hold up. Investors aren't just grading a quarter anymore–they're grading a thesis, one that a lot of very smart money has already bet trillions of dollars on being true.
My friends, you better pay attention tonight, because whatever happens after that bell rings isn't staying contained to two stock tickers. It's going to set the tone for how this whole AI trade gets treated for the rest of the summer. And Intel steps up later this week to answer the same question all over again–with an even higher bar. Stay hydrated. 🥤
YESTERDAY’S MARKETS
Stocks rallied yesterday, with the Dow closing at 52,224 (+0.7%), the S&P 500 gained 0.9%, and the Nasdaq climbed by +1.3%, led by a rebound in chip stocks including NVIDIA and Micron after the prior week's selloff. Brent crude climbed toward $92 a barrel–its highest level in five weeks–as attacks continued in the Iran war, up from under $72 earlier this month. The oil spike pushed Treasury yields higher.
NEXT UP
No major economic releases today.
Important earnings releases: WABCO, Phillip Morris, GE Vernova, Otis Worldwide, AT&T, Stifel Financial, IBM, Wyndham Hotels, Knight-Swift, ServiceNow, United Rentals, Las Vegas Sands, Texas Instruments, Tesla, Alphabet, Southwest Airlines, and Kinder Morgan.