Nvidia spent $58.1bn on other companies, nearly 12x last year's pace
The nine-month figure, pulled from Tuesday's results, nearly matched the $61.2bn Nvidia spent on its own buybacks, and it recasts the AI giant as an allocator.

Nvidia spent $58.1bn buying shares in other companies in the nine months to 25 October, almost 12 times the $4.95bn of a year earlier, and nearly matched the $61.2bn it spent repurchasing its own stock, Fortune found in the quarterly results posted Tuesday. For CFOs and boards, the figures mark a shift from a buyback-and-ship-it playbook to one where the chip supplier is also an owner across the ecosystem it sells into.
Nvidia spent $58.1bn buying shares in other companies in the nine months to 25 October. That is almost 12 times the $4.95bn it put to work the same way a year earlier, and Fortune spotted the line in the quarterly results Nvidia posted on Tuesday. Put that number beside the other big one in the same filing and the story sharpens: Nvidia spent $61.2bn buying back its own stock over the same stretch, up from $36.3bn. In plain terms, the company that supplies the picks and shovels of the AI boom is now writing two checks of roughly equal size, one to itself and one to everybody else.
The near-match is the headline. A buyback is a capital-return decision: it shrinks the share count, flatters per-share earnings, and signals that management sees its own stock as the best available asset. A $58.1bn program of stakes in other companies is the opposite kind of bet. It says the best available assets sit outside the building, and that Nvidia would rather own a slice of them than simply sell to them. Neither line is filed as a strategy document, but together they describe how a company with enormous cash generation is choosing to deploy it, and how fast that choice changed in twelve months.
The disclosure covers nine months, not a single quarter, and it lands as a lump. On its own it does not say which companies received the money, at what valuations, or whether the positions are passive or strategic beachheads. Historically, when a company of this scale buys minority stakes, the targets tend to sit close to its own value chain: chip designers and toolmakers, cloud and data-center operators, and the model builders that consume its hardware. That is the typical pattern for a supplier investing upstream and downstream at once, and it is why this number lands differently from a routine repurchase. One is a vote of confidence in your own share price. The other is a map of where you think the next decade of demand gets built.
There is a second-order question that follows any supplier that funds its own customers. When a chipmaker takes equity in the companies buying its chips, cash can, in effect, travel in a circle: out as an investment, back as revenue. Investors have historically been alert to this structure because it can make demand look sturdier than it is. It is not proof of anything, and the source figures do not break the total down far enough to support a conclusion either way. But it is the reason a $58.1bn investment line deserves more scrutiny than a $58.1bn buyback. Buybacks do not show up in anybody else's revenue.
Regulators add another layer. Minority stakes have historically sat below the thresholds that trigger merger review, which is part of what makes them attractive: you get influence and information without filing for approval. In recent years, competition authorities in the US and Europe have signaled more interest in exactly this kind of deal, particularly where an incumbent takes positions across an ecosystem it also supplies. Nothing in Tuesday's results suggests a challenge is coming. But a company that scales minority investing from $4.95bn to $58.1bn in a year moves from occasional investor to systemic one, and systemic investors tend to attract systemic attention.
The buyback line matters too. Spending on its own shares rose from $36.3bn to $61.2bn, an increase of roughly 69% on the prior-year figure. Buybacks are the default answer for a cash-rich company with few places to put money, and the easiest thing for a board to approve because they require no thesis about somebody else's business. The comparison is the point: external investing grew almost twelvefold while the buyback grew far more slowly. Nvidia is still returning enormous capital to shareholders. It is also, for the first time at this scale, behaving like an allocator with a view on other people's businesses.
Watch the next filing. A nine-month total sets a pace, and pace is what matters here: if the run rate holds, Nvidia's annual spending on other companies' shares would sit at a level few corporate balance sheets could match. Watch the disclosure itself, too. The more material these stakes become, the harder they are to keep inside a single line item, and the more pressure there will be to name names, mark positions, and explain losses as well as gains. Stakes are not free money; they carry mark-to-market risk that buybacks never do.
For CFOs and boards at other cash-rich tech companies, the takeaway is structural. Capital allocation is no longer a two-way choice between buybacks and dividends; the third option, buying equity in the ecosystem you sell into, is now being run at a scale that will be studied and copied. Startup founders and their investors should read it as a signal about who might appear in their next round, and on what terms. Competitors should read it as a moat built with balance sheet rather than product. And shareholders should read it as a question worth asking on the next earnings call: what exactly did $58.1bn buy, and who owns the upside?
This story's Key Insights and Take-aways are locked.
Create a free account to unlock Executive Actions for one credit.
Register to UnlockAlways free for Executives Club members. Join the Club
More in Business
S&P 500 Enters Correction, Down 10% From Its High on AI Worries
The index's slide past the 10 percent line is a symbolic verdict on investor mood, and the AI trade that powered the rally is now the thing testing it.
Dow and S&P 500 enter correction as Nasdaq logs worst week since 2020
All 11 S&P 500 sectors fell more than 1% in the same week for the first time since April 2025, a breadth signal that matters more than any single index level.
Chip stocks lose $2.3 trillion in a week, worst run in 25 years
A $2.3 trillion weekly wipeout in semiconductors is forcing every board with an AI story to ask whether it owns a thesis or just a crowded trade.



