2026-10-19

Deutsche Bank and Cantor Fitzgerald Call CoreWeave's Debt Panic Overdone - Price Targets Imply 100%+ Upside Even as Bernstein Stays Bearish

Written & reviewed by the TradeSmrt editorial team · 2026-10-19

In this article

What Happened
Why Bond Investors and Equity Analysts Are Reading the Same Company So Differently
What to Take Away From This
FAQ
Sources

What Happened

As this site covered earlier this week, CoreWeave's (Nasdaq: CRWV) stock plunged roughly 15%-16% on Wednesday, October 14, after yields on its unsecured bonds spiked past 15% and five-year credit-default swaps hit record distress levels near 1,140-1,195 basis points. The scare, triggered by a viral social-media exchange between investor Chamath Palihapitiya and Elon Musk, spread to Nebius Group (Nasdaq: NBIS), which fell as much as 9.7%, and even touched mega-cap Oracle (NYSE: ORCL). CEO Michael Intrator pushed back in an internal memo reviewed by Reuters, calling the reaction "irrational" and pointing to a $104.2 billion revenue backlog that he said "more than supports" the company's roughly $52 billion in debt and lease obligations.

What happened in the days right after is the part worth focusing on now, because it cuts against the panic narrative. Rather than piling onto the bearish case, several of Wall Street's largest bank research desks did the opposite. On Thursday, October 15, Deutsche Bank reiterated its Buy rating and $150 price target on CoreWeave - a target it had actually set back in August, left completely unchanged by the crash - which against the prior day's $65.78 close implied roughly 128% upside. The bank's analysts said they found no evidence of company-specific fundamental damage and pointed to CoreWeave's continued infrastructure partnerships with Nvidia, Dell, Meta, and OpenAI as evidence that underlying demand for its compute capacity remained intact. Cantor Fitzgerald went further in tone, reiterating an Overweight rating and a $176 target - more than 160% above the post-selloff price - and explicitly calling the selloff an attractive entry point, noting that nearly all of CoreWeave's 2027 capacity is already contracted and that Nvidia itself has agreed to backstop any unsold capacity. Citigroup kept a Buy rating too, though it formally tagged CoreWeave as "High Risk" alongside a $135 target.

Across the roughly 25-30 analysts who cover the stock, the consensus 12-month price target sat around $141 - implying well over 100% upside from where shares closed after the crash. That consensus wasn't unanimous, though: Bernstein's Madison Rezaei kept an Underperform rating and just a $74 target, arguing the uncertainty around AI infrastructure financing remained too large to dismiss. Retail money moved too. Cathie Wood's ARK Invest had already been adding to its CoreWeave position through earlier dips in the stock, and sentiment-tracking site Stocktwits noted CRWV and NBIS both edging higher in after-hours trading as bargain-hunters stepped in following the selloff. Separately, boutique research shop Rosenblatt initiated coverage of Nebius with a Buy rating and a $304 target, arguing the company's contracted AI capacity and infrastructure demand justified a bullish stance even with CoreWeave's credit story still unresolved.

Why Bond Investors and Equity Analysts Are Reading the Same Company So Differently

The split here isn't really a contradiction - it reflects two different markets asking two different questions about the same company. Bond and CDS markets are, by construction, asymmetric bets: a bondholder's best-case outcome is simply getting paid back in full, while the downside is losing most or all of their principal in a default. That structure makes credit markets hypersensitive to tail risk - the probability, however small, that a highly leveraged company can't refinance its debt when a loan or bond matures. Equity analysts at banks like Deutsche Bank and Cantor Fitzgerald, by contrast, are pricing a much wider range of outcomes, including the upside case where CoreWeave's committed backlog converts into real cash flow over the next several years. When a stock has already priced in enough fear - as a 15% one-day drop on top of weeks of prior declines arguably does - even a modestly reassuring set of facts can make the equity case look favorable, even while the credit case stays genuinely risky.

The mechanism behind CoreWeave's specific vulnerability is worth spelling out, because it explains why even bullish analysts aren't dismissing the story outright. The company has financed much of its GPU-data-center buildout with debt secured against long-term contracts with hyperscale customers - a structure that works well as long as those customers keep paying and as long as CoreWeave can keep refinancing maturing debt at reasonable rates. But once yields on its existing bonds spike past 14%, any new debt CoreWeave issues going forward gets priced off that elevated curve, raising its future cost of capital regardless of whether existing contracts perform exactly as promised. That's the specific point Seaport Research's analyst Jay Goldberg raised in a notably more cautious take than the big banks offered: CoreWeave and Nebius will both need to raise additional debt next year to keep building, which makes it hard to justify an outright upgrade even if he suspects the selloff overshot - in his words, there remains "just too much uncertainty" baked into the capital structure. Backlog revenue, as several analysts covering the story have pointed out, also isn't the same thing as cash in hand; it's contracted future business that still depends on data centers getting built, powered, and delivered on schedule.

This tug-of-war between credit caution and equity optimism landed in the middle of an already rough week for the broader market. The Nasdaq Composite fell 4.5% for the week ending Friday, October 16 - its worst weekly performance since June, as the Philadelphia Semiconductor Index logged its ugliest week since April 2025 - while the S&P 500 closed Friday at 7,659.44, down 1.09% on the day. The Dow Jones Industrial Average, with far less AI exposure among its 30 components, actually rose about 0.7% over that same stretch. The CBOE Volatility Index touched a four-month intraday high on Friday as the 10-year Treasury yield hovered near 5.10%, while West Texas Intermediate crude jumped 3.5% to settle near $101.41 a barrel on fresh U.S.-Iran and Russia-Ukraine tensions - an additional source of nerves layered on top of the AI-credit story.

Heading into Monday, October 19, futures pointed to at least a partial stabilization: S&P 500 futures were indicated roughly 0.4% higher and Nasdaq 100 futures around 0.6% higher in premarket trading, as investors shifted attention toward a dense week of earnings. That matters because the underlying earnings backdrop has, so far, actually been solid: among S&P 500 companies that have already reported third-quarter results, roughly 85% have beaten analyst estimates, with an average positive surprise of about 3.3%. In other words, the credit-market anxiety concentrated in a handful of heavily leveraged AI infrastructure names has been running on a largely separate track from the broader corporate earnings picture - which is exactly the kind of divergence that can make headline index moves misleading if you don't look at what's actually driving the number underneath.

What to Take Away From This

A bond market signal and an equity analyst's price target can disagree sharply on the same company, and both can be internally consistent. Credit markets price the risk of catastrophic loss; equity research prices a range of plausible outcomes including recovery. Seeing a stock's bonds trade near distressed levels while its equity analysts reiterate Buy ratings and triple-digit upside targets isn't a contradiction you need to resolve in one direction - it's useful information about where the real uncertainty sits.
A reiterated price target that survives a crash unchanged is a stronger signal than a target set after the fact. Deutsche Bank's $150 target predated the October 14 selloff by two months; the bank didn't need to raise it to generate a large implied upside, which suggests conviction rather than a reflexive post-crash bounce call.
Revenue "backlog" is not cash in the bank, no matter how large the number. When a company under financial scrutiny points to a large contracted-revenue figure to reassure investors, it's worth asking how much of that revenue depends on future execution - building facilities, securing power, delivering service - rather than being money already collected.

FAQ

Why would big banks keep Buy ratings on a stock whose bonds are trading like distressed debt?

Because equity and credit analysts are typically different teams asking different questions. Equity researchers at Deutsche Bank and Cantor Fitzgerald focus on whether CoreWeave's contracted revenue will eventually turn into durable profit, while the bond market is pricing the narrower risk of whether the company can keep refinancing debt smoothly along the way. A bank can genuinely believe both that the equity has room to run and that the credit situation deserves a "High Risk" label, the way Citigroup did.

If multiple banks say CoreWeave has 100%+ upside, does that mean the selloff is over?

Not necessarily. A price target reflects one bank's model of where a stock could trade over roughly 12 months under its own assumptions, not a guarantee. Bernstein's Underperform rating and $74 target - a fraction of the bull-case numbers - shows professional analysts still disagree meaningfully about the outcome, and the credit market's distress signals hadn't fully reversed in the days right after the crash.

Is Nebius facing the exact same risk as CoreWeave?

Related but not identical. Nebius fell in sympathy with CoreWeave because both are seen as financing AI infrastructure buildouts with significant leverage, and investors treated the October 14 scare as a sector-wide story rather than a company-specific one. But Rosenblatt's decision to initiate Nebius specifically at Buy, even while CoreWeave's credit situation remained unresolved, suggests at least one research shop sees meaningful differences in how the two companies' contract structures and balance sheets are built.

Sources

This article synthesizes and analyzes reporting from the sources below; it does not reproduce their original text.

This article is provided for informational purposes only and does not constitute a recommendation to buy or sell any security. Investment decisions and their outcomes are solely the responsibility of the investor.