AI is becoming a financial engineering business
Fortune Amit Joshi ● Covered by 4 sources
AI is turning into a balance-sheet race, not just a model race. Big Tech is spending trillions, and the winners may be the firms that can finance the most compute.
Based on reporting by Fortune, Amit Joshi — read the original for the full story.
Summary, retelling and take written by AI under human oversight; images are AI-generated illustrations. How we work · Report an error
AI’s big surprise isn’t that the models got better. It’s that they’re starting to look replaceable. Once that happens, the moat shifts away from who built the smartest chatbot and toward who can afford the data centers, chips, power and financing to keep the whole machine running.
That shift is already brutal in size. Since the AI boom began in 2023, Amazon, Microsoft, Alphabet and Meta have together spent $1.1 trillion on AI infrastructure. They plan to add another $745 billion this year alone. This is a very different business from the asset-light software era investors loved for the past 20 years. Capital intensity has come roaring back, and there’s no polite way to avoid it.
The financing side is getting almost as important as the engineering side. Nvidia is working with Apollo, Blackstone, Goldman Sachs and others to mobilize more than $500 billion in extra capital for AI infrastructure. Google has built a $200 billion financing structure with Broadcom, Apollo, Blackstone and Morgan Stanley to fund Anthropic’s chips and data centers. When the lenders, the chipmakers and the cloud vendors all crowd into the same trade, the industry starts to look less like software and more like a very expensive infrastructure market.
The money is not flowing into a vacuum, though. Microsoft’s cloud business grew 32% to $39.3 billion in the latest quarter, while Amazon Web Services rose 37% to $42.2 billion and Google’s cloud business jumped 82% to $24.8 billion. Microsoft’s overall revenue rose 18%, and investors added a record $450 billion to its market value in a single day after the results. So the spending is finding real demand, at least for now.
But the pressure on balance sheets is real. Alphabet’s free cash flow has gone negative for the first time since its IPO, and Meta’s free cash flow also fell sharply in the latest quarter. IBM’s second-quarter results showed a different kind of pain: customers delayed software purchases as they rushed to secure AI infrastructure ahead of expected price hikes, and the stock dropped 25% in a day in mid-July. Apple, by holding back, briefly got treated like the sensible adult in the room with a $5 trillion valuation last month. The market is clearly split. One camp is betting on scale and spending; the other on restraint. And that’s the point: AI is no longer just a model contest. It’s a financial engineering contest.
My take — AI-written commentary, not fact-checked reporting
The hype crowd keeps talking about intelligence, but the real action is on the funding table. That should make anyone who loves “software margins” a little nervous, because the new moat looks an awful lot like debt capacity, power access and who can keep the servers humming without blinking. Open or closed matters less than who can afford the bill.
Read more about this at: Fortune