Morgan Stanley initiated credit coverage of Nvidia this week, and its verdict landed with more hedging than Wall Street's usual buy-side enthusiasm: the bank's model and relative-value analysis are "supportive," but the long-term risk in Nvidia's financing structure is still too hard to measure. "Our model and relative value are supportive, but we think the tail remains too early-stage, opaque, and sizable to step in," Morgan Stanley credit analyst Lindsay Tyler wrote in the note, issued Monday. [1]
The timing matters. This paper covered Wednesday's earnings under the headline Nvidia Beats Estimates by $4 Billion and Guides to 70 Percent Growth, reporting that the same quarterly filing which produced a clean beat-and-raise also disclosed, for the first time, "indebtedness" as a standalone risk factor — with debt due within one to five years jumping from $2.75 billion to $15 billion in a single quarter. [2] Morgan Stanley's note arrived one day before that print, meaning the bank's credit desk flagged the opacity of Nvidia's financing tail before the company's own 10-Q made the debt disclosure explicit. The two documents, read together, tell the same story from different vantage points: the company that is borrowing has said so in a filing, and the bank pricing that debt has said it cannot yet size the risk.
Nvidia's own numbers explain what Morgan Stanley is trying to assess. The company disclosed $33.5 billion in senior notes outstanding and an active $25 billion commercial paper program — the residue of what CNBC described as Nvidia's first bond sale since the AI boom began. [1] Supply commitments, chiefly memory procurement, more than doubled quarter over quarter, from $119 billion to $279 billion. [1] Set against that is a financing initiative chief executive Jensen Huang announced earlier this month: a group of six major financial firms pledging up to $500 billion based on the thesis that AI chips are themselves an investable asset, letting Nvidia offer residual-value backstops on projects built around its hardware. [1] Morgan Stanley's note is effectively a credit-market referendum on that thesis, arriving at a qualified "not yet."
CNBC's own live coverage of Wednesday's earnings treated the Morgan Stanley note as one item among dozens in a scrolling blog dominated by beat-and-raise numbers — $96.22 billion in quarterly revenue against a $92.17 billion consensus, and guidance for 70 percent revenue growth in fiscal 2028 versus the 44 percent analysts had modeled. [1] The credit note appeared far down the page, positioned as color rather than headline. That placement is itself part of the story: a bank's credit analyst calling a major financing structure's tail risk "opaque" is, in ordinary circumstances, treated as newsworthy on its own terms. Buried under an unambiguously strong earnings report, it read instead as a footnote to a rally.
The credit note is not Morgan Stanley's only recent signal on Nvidia's financing architecture. Huang's $500 billion financing initiative rests on a specific thesis — that AI chips have become an investable asset class in their own right, durable enough for major financial firms to offer or accept as collateral. Morgan Stanley's note engages that thesis directly rather than dismissing it: the bank's own model and relative-value work back Nvidia's credit story, which is a meaningfully different posture than a downgrade or a sell recommendation would represent. What Tyler's note withholds is confidence in the tail — the low-probability, high-severity scenario in which financing structures built around still-unproven long-term chip demand curves come under stress simultaneously across the industry. That is a narrower and more technical objection than "the AI boom is a bubble," and it is also harder to dismiss, because it comes from the desk whose job is specifically to price that kind of risk for institutional clients, not to generate headlines.
What makes the note notable is who is saying it and in what capacity. Short-sellers and skeptical newsletter writers have questioned Nvidia's financing arrangements for months; this paper's Wednesday coverage noted finance commentators on X, including Ed Zitron and the Kobeissi Letter, arguing the company's growth increasingly runs on borrowed capital rather than organic cash generation. [2] Morgan Stanley is not a short-seller. It is a bank initiating formal credit coverage — the kind of research that informs how other institutions price Nvidia's debt — and its analyst reached a similar conclusion through an entirely different discipline: not equity valuation, but credit risk. That two separate vantage points, bearish commentary and a bank's own credit desk, converged on the same word — opaque — is the detail Wednesday's beat-and-raise coverage did not have room to sit with.
-- THEO KAPLAN, San Francisco