Big Tech's Cash Is Draining. Its AI Business Is Earning More

Nomura's J-curve points to a timing gap between falling free cash flow and the payback on AI spending. For investors, the two are not the same thing.

The 20-second read

1
Profits from the AI business are growing. In the third quarter of 2026, AWS revenue rose 39% from a year earlier and Google Cloud revenue rose 103%. Cloud operating margins improved at both companies.
2
Cash flow, however, is moving the other way. Amazon's trailing 12-month free cash flow is -$42.4 billion, and Alphabet's third-quarter FCF is -$9.1 billion. Nomura interprets this as a J-curve effect, in which huge infrastructure spending is paid out first.
3
What investors need to check is the ability to bring cash back in. Cloud revenue and margins, contract backlog, and the growth rate of capital expenditure should be viewed together. Whether the J-curve actually recovers depends on the pace of future investment and cash generation.

Is cash falling because the business isn't making money?

Amazon generated $168.3 billion in operating cash flow over the past 12 months. Its free cash flow was still negative, at $42.4 billion. Capital spending, led by AI infrastructure, surged.

Alphabet tells a similar story. It reported third-quarter operating cash flow of $42.3 billion, but capital expenditure reached $51.4 billion. Free cash flow (FCF) fell to -$9.1 billion.

When a company earns cash from operations but spends more than that on data centers and chips, FCF turns negative.

The point that matters is whether a deficit comes from weak operations or from expanded investment for future earnings.

The cloud units are already highly profitable

Metric (Q3 2026, YoY)AWSGoogle Cloud
Revenue growth+39%+103%
Operating margin38.7%36.2%

Source: Amazon and Alphabet third-quarter 2026 earnings releases.

AWS posted quarterly revenue of $46 billion and operating income of about $17.8 billion. Google Cloud posted revenue of $30.7 billion and operating income of about $11.1 billion.

Google Cloud's operating margin rose to 36.2% from 23.7% a year earlier, a gain of 12.5 percentage points.

That suggests cloud profitability can improve even as AI infrastructure spending grows. One caveat: the cloud segments also include legacy businesses, so these figures do not directly prove the return on new AI data center investment.

What the J-curve really means: cash goes out first, comes back later

Nomura's J-curve is simple.

Building a data center and buying GPUs sends out a huge amount of cash at once. The revenue and cash a data center generates once it is running arrive over several years.

Companies also build new data centers every year. Cash may start flowing in from last year's facilities, but if this year's investment is larger, overall FCF can keep worsening.

How the J-curve works

A conceptual illustration of the J-curve principle: cash falls during the initial investment burden, then rises as investment is recovered. It does not show actual company FCF figures or Nomura's estimated curve.

Nomura's Nov. 2 note suggests there may be a lag between operating profit growth and an FCF recovery at Big Tech. It models the cash flow burden around 2028 and the recovery path after that.

There is an important condition. Even if investment keeps rising, the J-curve recovers only if cash generated by existing facilities grows faster than the burden of new investment.

Customers are waiting. The question is contract quality

The numbers behind the AI supply shortage are large.

CompanyDisclosed contract balance
Microsoft commercial RPO$677 billion
AWS backlog$578 billion
Google Cloud backlog$560.8 billion

Microsoft's commercial remaining performance obligation (RPO) rose 73% from a year earlier. AWS disclosed its $578 billion backlog in its third-quarter earnings discussion. Google Cloud's is $560.8 billion.

Contracts worth hundreds of billions of dollars are a positive sign of long-term demand. But a contract balance is not cash inflow. The speed at which it converts to revenue depends on customer usage, contract length, cancellation terms and the customer's ability to perform.

A key issue is whether AI company customers earn enough to keep paying for massive amounts of infrastructure.

The bottom line: the ability to recover cash after investing matters as much as the size of the AI spending itself. Investors who read the cause behind the minus sign on the cash flow statement, and the path to recovery, are likely to judge better than those who read only the sign.

Can operating profit be positive when FCF is negative?
  • Yes. Operating profit reflects depreciation spread over time, while the cost of buying equipment is recorded when the cash goes out.
If it is a J-curve, will recovery come without fail in 2028 to 2030?
  • No. That path is Nomura's estimated scenario. The timing of recovery can change depending on the growth rate of new investment, the profitability of existing equipment, and customer demand.
Are semiconductors better positioned than hyperscalers right now?
  • During an infrastructure expansion, chip suppliers can benefit first. But when investment slows, earnings volatility at chip companies can grow larger. Valuation also needs to be weighed.
What is the warning sign investors should watch most closely?
  • It is a situation where investment keeps rising while revenue growth and margins slow at the same time. That could signal a worsening return on investment, not just a timing gap in cash flow.

What to watch

Cloud growth and operating marginWhether growth holds up while margins improve.
CapEx growth and operating cash flowWhether the pace of investment growth is slowing relative to the pace of cash generation.
Conversion of RPO and backlog into revenueWhether contracted demand turns into actual billings and cash receipts.
GPU utilization and rental pricesWhether the supply shortage persists, or early signs of overcapacity appear.
Leases, borrowing, and share issuanceWhat cost the financing needed to sustain investment imposes on shareholder value.
Big Tech's cash deficits likely reflect AI investment more than weak operations, and the question is when and how much of that money comes back.

Insight Times Editorial Desk

More by topic

Starting with this story's topics, then what readers are following now

Investor psychology, decision-making and books on investing

CapEx

This story's topic71 storiesSee all →

Money a company spends on things it will use for years — plants, equipment, bu…

Oracle Sets a Date for AI Payback: $30 Billion in Free Cash Flow by Fiscal 2029

Oracle laid out when its AI data center spending should start returning cash: positive free cash flow in the second half of fiscal 2028 and more than $30 billion in fiscal 2029. The plan only works if contracts hold and customers can pay.

Tech

Tech Selloff: Is the Bull Market Over? Wall Street's Four-Year Chart Says Not Yet

The uptrend has survived five double-digit pullbacks in four years. This time, AI companies' debt and cash flow are the new variables.

Investing

White House Swaps AI for "SI." The Real Investment Signal Is Infrastructure, Not the Name

An executive order renames artificial intelligence "Super Intelligence" and six frontier labs sign a voluntary safety accord. What investors should watch is the data center, power, nuclear and chip policy now gathered under one national-competitiveness frame.

Tech

AI Infrastructure Spending Is Holding Up 3% Global Growth, the IMF Says

AI is acting as a real-economy investment engine, offsetting an energy supply shock. The next test is whether that spending comes back as profit.

Investing

Morgan Stanley Says 'Buy the Drop' on AI Stocks, Keeps $3.4 Trillion Capex Forecast

After a post-midterm selloff on data center regulation fears, Morgan Stanley argues spending is likely delayed, not lost. The winners may differ, though.

Investing

Year Four of the AI Revolution: The Real Winners May Emerge After the Highway Is Built

Railroads, the internet and mobile show that a technology can succeed while many investors lose money. The question for AI is whether the current pullback is the end of the boom or a handoff of wealth to new owners.

Investing

Dark navy cover. Flow diagram: Nvidia to Nscale (credit guarantee, in talks) to Anthropic (~$45B compute lease signed in August plus a separate ~$45B, ~460MW deal in talks). Title: Why Would Nvidia Back a Customer's Loan?

Why Would Nvidia Guarantee Its Customers' Data Center Loans?

Nvidia is reportedly in talks to back financing for Nscale's West Virginia data center. Nothing is signed, and the terms are not public.

Tech

GOOGL

This story's topic29 storiesSee all →
Smartphone showing a Gemini trip widget on an airport cafe table

Google Wants Gemini to Build Your Phone Screen. The Real Prize Is the First Request

Create My Widget lets users describe a screen and have Gemini build it. For Alphabet investors, the bigger question is who owns the customer's first request, and usage and retention data should show it.

Tech

Same cloud, three different ways to earn the money back

Microsoft, Alphabet and Amazon are all spending enormous sums on AI infrastructure. The route each one takes to recover that capital is not the same, and that is what decides the winner.

Investing

The blue links are disappearing: what a website is worth when AI ends the search

Search is shifting from finding good pages to writing the answer itself. The convenience has a bill attached, and the people who produce the original pages are the ones paying it.

Tech

Nvidia and Google Want AI to Take a Break When the Grid Runs Short

The power problem for AI data centers has looked like a question of how many new plants to build. Nvidia and Google are offering a second answer: delay the AI work that can wait until the grid's worst hours pass.

Tech

Engineer reviewing Google DeepMind AlphaProof Code verification of Signal code

AI Can Write Code. Now Google DeepMind Is Using It to Prove the Code Is Safe

DeepMind's AlphaProof Code, working with human experts, mathematically verified key security properties of Signal's encryption code. The next AI coding contest may turn on how reliably code can be verified, not how fast it is written.

Tech

Why Oracle Is Buying More Nvidia GPUs and Adding Google TPUs Too

Oracle will expand Nvidia's next-generation AI systems to multi-gigawatt scale while becoming the first big outside cloud to offer Google's TPUs. The move shows competition shifting from single-chip speed to the running cost of whole systems.

Tech

AI Competition Shifts From Chatbot Share to the 'Action Layer'

ChatGPT isn't collapsing. But as Gemini, Claude and Copilot grow through their own distribution channels, the real contest is moving from standalone chatbots to an action layer wired into search, operating systems, workplace tools, ads and payments.

Tech

AMZN

This story's topic13 storiesSee all →
A person at a kitchen table looks at a food delivery app on a phone beside a laptop and a DoorDash delivery bag

AI Was Supposed to Do Our Shopping. DoorDash Got Only Several Hundred Orders From It

DoorDash logged 993 million orders in the third quarter, yet outside AI agents placed only several hundred in total. Orders built through its own in-app AI ran about 50% larger.

Tech

Amazon Goes After Nvidia's Strongest Moat, and It Is Software, Not GPUs

Amazon's open-source TorchNeuron aims to let existing AI code run on its Trainium chips. The contest comes down to the total cost of AI compute and the return on Amazon's $220 billion in annual capital spending.

Tech

The industry wins, the shareholder loses: AI's most awkward scenario

The biggest risk in AI is not that the technology fails. It is that AI works, spreads through the economy, and still destroys returns because too much capacity gets built. The internet succeeded. Plenty of dot-com investors did not.

Investing

Anthropic's IPO Filing Flags Two Risks: An AI That Found Passwords, and Revenue Routed Through Big Tech

Anthropic's S-1 describes an agent that left its test sandbox and a sales channel dominated by Amazon and Google. Both could shape costs and growth.

Tech

Amazon Targets Nvidia's CUDA Moat Before Its GPUs

The new front in AI chips is not raw speed but what it costs a developer to switch. Amazon's TorchNeuron aims to lower the barrier around the CUDA ecosystem Nvidia has built for more than a decade.

Tech

The Next AI Data Center Bottleneck May Be Local Consent, Not Power

Amazon is adding $1 billion for communities near its data centers. It signals that scarce AI inputs are shifting from GPUs and power to permits, grid access and local acceptance.

Tech

AI's Bottleneck Isn't One Chip Anymore

Huawei's AI accelerator prices jumped because of HBM costs, Amazon built a deal worth up to $60 billion with Qualcomm, and OpenAI pulled financial data and workflows into ChatGPT. Three different stories point to one shift: AI investment is no longer just about buying more GPUs.

Tech

FCF

This story's topic17 storiesSee all →

Why the White House Swapped "AI" for "Super Intelligence"

Executive Order 14434 renames AI across the federal government without defining a new technology. The real story is about roughly $700 billion in capex, free cash flow and electricity.

Tech

AI Demand Is Proven. Now the $70 Billion Bill Comes Due

Oracle posted $19.3 billion in quarterly revenue and spent $28.5 billion on capex. The real question for the AI buildout is no longer whether demand exists, but how fast this spending turns into cash.

Tech

CoreWeave's Bond Slide Is the First AI Warning, and It Came From Credit, Not Stocks

CoreWeave's 2032 notes fell below 80 cents on the dollar, with yields near 15%. Why is the bond market flashing a warning at a company swamped with AI demand?

Tech

Lam Research Falls 12%: The Problem Is the Timing of AI Growth, Not Its End

Revenue up 54%, an operating margin above 40% and $3.2 billion in free cash flow were not enough. The selloff likely reflects worry that revenue arrives later than hoped.

Investing

If AI Is Holding Up the Global Economy, Why Is the IMF Warning of an AI-Driven Financial Crisis?

AI succeeding and AI investors succeeding are different things. The IMF's concern is less that the technology fails than that the huge capital poured into it is not recovered.

Tech

Dark navy cover. Horizontal bar chart of Reddit Q3 2026: revenue +53%, ad revenue +54%, global DAU +12%, US DAU -2%, stock about -29% on Oct 30. Title: Google AI Is Taking Reddit's Visitors.

Google's AI Search Is Cutting Reddit's Traffic. Are the Web's Economics Changing?

Reddit's revenue grew fast, but search visits fell. The shock points to a shifting relationship between companies that produce content and platforms that answer with AI.

Tech

Google Cloud Grew 103%. Alphabet Fell 10% Anyway. Here Is What Wall Street Fears

AI demand is showing up in the numbers, but search ads missed and AI spending is not yet returning enough cash. Alphabet's results ask not how fast it is growing, but how profitably.

Investing