Amazon's AWS boom isn't enough when free cash flow turns negative

What does it take for a company to grow its cloud unit at the fastest pace in nearly five years and still get punished? Apparently, spending $77.3 billion in a single quarter.

That is what happened to Amazon on Thursday. Net sales rose 12% to $201.9 billion, and AWS climbed 39% to about $46 billion, with a record operating margin of 13.9%, according to Amazon's results as reported by Newsroom America. Yet earnings per share of $1.89 missed the $1.97 consensus, fourth-quarter revenue guidance came in below expectations, and shares fell in after-hours trading. Reported drops ranged from more than 5% to as much as 11%, depending on the outlet and the moment.

The number that mattered wasn't growth

The headline was AWS, but the line that moved sentiment was cash. As CNBC reports, capex hit $77.3 billion in the quarter, against $35.1 billion a year earlier and well above the $58.7 billion analysts expected. Trailing twelve-month free cash flow was an outflow of $42.4 billion, after an inflow of $14.8 billion a year before.

Put simply: operating cash flow rose 29% to $168.3 billion over the trailing twelve months, but Amazon is pouring even more into servers, chips and data centers. It expects about $220 billion of capex in 2026 and says 2027 will be "meaningfully more."

In our view, that is the real story. A 39% growth rate is impressive, but growth bought with a widening cash deficit is a different product from growth that funds itself. The market is not saying AWS is failing. It is saying the bill has arrived before the receipt.

Management's case is decent, but it's still a promise

To be fair, Amazon's argument isn't weak. Backlog is $578 billion, according to TheWrap, and Andy Jassy says AWS can grow above 40% in both 2027 and 2028. He also says operating margins on the AI side are "well ahead" of where the core business was at a similar stage. AWS operating income rose 55% to $17.77 billion, which suggests the spending isn't being burned for nothing.

But note the tense: projections, backlogs and a "clear line of sight to attractive financial returns" are forecasts. Backlog is contracted demand, not collected cash. Our read is that investors have simply run out of patience for forecasts at this scale. Peers are sending mixed signals too: Microsoft trimmed its capex outlook to $170 billion, while Alphabet held its guidance at $195 billion to $205 billion. When one hyperscaler cuts and another doesn't, the debate is no longer whether to spend but how to prove the return.

There is also a quieter weakness. Online sales grew just 2% and advertising missed estimates, partly because Prime Day moved into the second quarter. That timing effect is real, but it also means the retail side isn't cushioning the AI bet right now.

For savers, this is a concentration lesson, not a trading signal

We are not suggesting anyone buy or sell Amazon. But anyone holding a broad index fund should read this episode carefully. A handful of giant companies are all making the same enormous bet, and indices are weighted by size. So your exposure to "AI capex converts into cash" is probably larger than you think, even if you never bought a single tech share directly.

That isn't a reason to panic. It is a reason to check how much of your fund sits in the biggest names, whether your other holdings genuinely diversify it, and whether your time horizon can absorb a bad stretch if the market reprices these bets. Rules and tax treatment differ by country, so any rebalancing should reflect your own situation.

Our verdict: Amazon's quarter was operationally strong and financially uncomfortable, and both are true. The market's reaction is a reminder that, at $220 billion a year, "it's working" has to eventually show up as free cash flow. Until it does, expect volatility every time a hyperscaler reports, and treat a concentrated index as the risk it is, not the safe default it feels like.