Joseph Carlson — Market Sentiment Is Quickly Changing

Amazon's cloud business, AWS, rents computing power and storage to other companies. For years the market worried that Amazon was pouring enormous sums into data centres with nothing to show for it. This quarter answered that: AWS revenue grew 36.7% from a year ago to $42.2 billion — the fifth quarter in a row where growth got faster, and the fastest in four and a half years. Annualised, that one division now runs at $169 billion a year: more than three times Netflix's entire revenue, about 70% bigger than all of Tesla, and roughly half of Berkshire Hathaway — which, at this pace, it passes around 2029.

Two details matter more than the headline. First, the backlog — money customers have already contractually committed to spend — grew 154% to $496 billion, faster than revenue itself, and Amazon says the capacity being built through 2028 is already spoken for. That is demand you can see two years out rather than forecast. Second, it is getting more profitable while growing this fast: $16.6 billion of operating profit at roughly 39% margins, helped by its own custom chips and networking gear. Growth, visibility and profitability all beat at once, which is why CEO Andy Jassy's claim that AWS becomes a trillion-dollar-revenue business gets taken seriously.

Carlson called this reacceleration in September 2025 and says he still lowballed it — "I underestimated it to a huge extent." Amazon is now $207,000 in his Story Fund, up $82,000, and deliberately one of his two overweight positions alongside Google.

Microsoft is the other big cloud spender the market had turned on. Its report was strong enough to move a $3-trillion-plus company up 16% in a single day — then another 5% — wiping out every loss it had taken in 2026. Carlson holds it (an $80,000 position, up $37,000) and counts the report as part of what flipped sentiment.

His nuance is about causation: the rest of the group only moved after Amazon reported, not after Microsoft. So Microsoft helped, but Amazon is what changed investors' minds about heavy AI spending generally.

Google put out what Carlson calls an incredible earnings report and investors sold it anyway — the same pattern as Meta. Now that Amazon has shown what the heavy AI spending actually buys, Google is getting a lift too (up 3% on the day) and his position is back to $70,000 of gains.

At $223,000 it remains his single largest holding, in what he calls a neck-and-neck race with Amazon for the top spot. He is comfortable being deliberately overweight both — the concentration is intentional, not accidental.

Meta is the name investors most disliked in this earnings round; they sold it after the report. It is recovering as the AI-spending fear lifts across the group — up 5.5% on the day — and Carlson's loss on it has narrowed from more than $30,000 to $24,000 (roughly 14% down).

He does not re-argue the case here, but he is holding through the drawdown, consistent with buying more the week before. The read-across is the point: the same evidence that re-rated Amazon and Microsoft is what pulls Meta back up.

Mastercard runs the network that moves money between banks whenever a card is used. It is a long-standing core holding, and here it simply shows up as working — the first name he points to when his portfolio pushes toward all-time highs, having "moved up big over just the past couple of months."

S&P Global rates the creditworthiness of bonds, owns the S&P 500 and Dow Jones indexes, and sells financial data by subscription. Carlson presents it as the answer to the FICO problem: same broad industry, but the revenue comes from several independent engines rather than one product, so no single regulatory or competitive shock can break it.

The quarter was quietly good and getting better: revenue up 11% once you strip out businesses it has spun off (Mobility Global), an "incredible" ratings business, growing indexes, and Market Intelligence — the piece investors watch most — up 6% with 9% more customer volume. Because the cost base barely moves, an 11% revenue increase turns into much faster earnings-per-share growth. He owns it and stayed bullish on the report.

Uber is one half of a "split buy" he made a month ago — an equal dollar amount into Uber and DoorDash at the same moment, treating them as a paired bet on delivery and mobility networks. It is down about 2% so far and reports this week.

He concedes Uber loses US food delivery to DoorDash, but thinks it can dominate European delivery through acquisitions, and that ride-hailing itself is a far bigger prize. On the robotaxi fear, his argument is by analogy: nobody says Visa is doomed because Mastercard, Amex, PayPal and Venmo exist — a dominant network doesn't need 100% of a market to be enormously profitable. And if self-driving cars work, Uber can simply buy or license them and run them in its own fleet rather than being displaced.

He is explicit that this is higher risk than his core holdings — regulation, competition, and a bumpy path. Into earnings he is watching rides and Uber One membership growth, not profits: the company is still in land-grab mode.

DoorDash is the other half of the split buy, already up 15% (+$3,000), reporting this week. The core fact is share: around 60% of US food delivery today, and expected to move toward 70% — a lead that compounds into what he calls "a dominant unbreakable moat."

His demand argument is that paying for delivery still strikes many people as a waste of money, but that this changes structurally: as incomes rise and the cost of delivery falls, more people buy convenience — and "DoorDash represents convenience." Into the print he cares about delivery volumes and DashPass subscriptions rather than profit, because the company is still scaling.

Duolingo is the language-learning app, and it is his worst position — a $33,000 stake sitting $13,400 in the red. He is holding every share into this week's earnings, and uses it to explain the "asymmetry of stocks": Peter Lynch only had about six winners in ten, but the winners rose far more than the losers fell. Carlson's own version is stark — his single biggest mistake has cost $13,000 against more than $500,000 of lifetime gains. A portfolio should never be able to collapse on one name.

He is unusually candid that this is not a high-confidence call: "I'm not 100% confident of Duolingo." It is a small, volatile company with many bears, and even after the fall it trades at a valuation that assumes solid growth and improving profitability — so there is real downside if the numbers stay weak.

What keeps him in is the product. Learning a language on Duolingo used to be typing and filling in blanks with almost no speaking; AI has turned it into constant spoken conversation with the app's characters, fixing its biggest long-standing flaw. Product improvements show up in the reported numbers only later — "the product improves and then the metrics will follow suit over time."

Texas Roadhouse is a steakhouse chain and one of his best-ever investments — a $61,000 position with $53,000 of gains, at all-time highs ($211, up 24% this year before dividends). He calls it a compounding machine and "the best operated" restaurant company in the world, with plenty of room to keep opening locations.

The part most people miss is where the next leg comes from. It is putting up these numbers while two of its biggest costs — beef and wages — are unusually expensive. Those costs are cyclical. When they normalise, the savings drop straight to the bottom line, which is why he sees "an incredible margin expansion story" ahead rather than a stock that has already had its run.

FICO owns the credit score used across US lending — about as dominant a position as a company can have — yet the stock is down 32% this year, from $2,300 to near $1,000, and fell another 7% on its latest report. Carlson has never owned it and explains why the dominance isn't enough.

Two things are happening. The first is pure valuation: the stock had reached 100 times earnings. That multiple is only survivable while investors are afraid of nothing about the future — the moment any fear appears, a 100× price cannot be defended. The second is that the fear is legitimate. The regulatory protection that guaranteed FICO's position has eroded, and it is now in an open commercial war with the credit bureaus that distribute its scores, which is hurting both sides.

His verdict is not that the business is broken — it still grows profitably — but that it is no longer predictable, and predictability is what the premium was paying for. The market now pays about 22 times forward earnings instead. Given the choice within the same industry, he owns S&P Global instead, because its revenue doesn't depend on any single thing going right.

Equifax is one of the credit bureaus that distributes FICO's scores — and now competes with FICO directly. Carlson's point is that this fight has no winner so far: "both companies like Equifax and FICO have gone down substantially as they battle it out." When a former partnership turns competitive, both sides typically spend more and earn less, which is why he isn't interested in either side of the trade.