Start with the plainest answer. CoreWeave said Q1 net loss widened to $740 million from $315 million a year earlier, while revenue rose 112% year over year to $2.1 billion, CoreWeave Q1 2026 Earnings Transcript reported this week. Revenue growth is doing its job. Earnings, not so much.

The spending side is what makes people flinch. CoreWeave said Q1 capex totaled $6.8 billion and raised full-year capex guidance to $31 billion to $35 billion, against 2026 revenue guidance of $12 billion to $13 billion, CoreWeave Q1 2026 Earnings Transcript reported this week. That is a huge gap. It can be justified if the infrastructure keeps filling quickly and margins rise as promised. It looks less charming if deployment slows or financing gets more expensive.

Debt is the other reason the market is uneasy. CoreWeave said Q2 interest expense is expected to land between $650 million and $730 million, up from $536 million in Q1, CoreWeave Q1 2026 Earnings Transcript reported this week. An outside analysis from my.stock.research estimated total debt at about $33 billion after April’s capital raises and put the current ratio at 0.46. That is the author’s estimate, not a CoreWeave disclosure. Still, it points in the same direction: this is not a balance sheet with a lot of spare tire.

Management’s reply is coherent, which is not the same thing as reassuring. CoreWeave said new deployments incur lease, power, and depreciation costs before they start generating revenue, a process that typically takes about one to two months, after which contribution margins normalize in the mid-20s, CoreWeave Q1 2026 Earnings Transcript reported this week. The company also said its adjusted operating margin was 1% in Q1 and expects that to be the low point as capacity ramps. Fine. But investors still have to believe the timing gap closes on schedule while CoreWeave keeps funding one of the fastest buildouts in public markets.

The bullish case does not need a lot of hand-waving. The numbers are already loud.

CoreWeave reported Q1 revenue of $2.1 billion, adjusted EBITDA of $1.2 billion, and a 56% adjusted EBITDA margin, CoreWeave Q1 2026 Earnings Transcript reported this week. It also said it signed more than $40 billion of new commitments in the quarter. That pushed backlog to $99.4 billion. A company does not usually accumulate that kind of order book by accident.

The growth curve is still extraordinary by infrastructure standards. LongYield Substack reported in March that CoreWeave went from $229 million in 2023 to $1.92 billion in 2024 to $5.13 billion in 2025, and CoreWeave itself said it expects full-year 2026 revenue of $12 billion to $13 billion. The company now expects to end 2026 with annualized run-rate revenue of $18 billion to $19 billion, and to exit 2027 above $30 billion, with more than 75% of that already contracted, CoreWeave Q1 2026 Earnings Transcript reported this week.

Backlog matters because it is not just wishful thinking stuffed into a slide deck. CoreWeave said 36% of the backlog should be recognized in the next 24 months and 75% within four years, CoreWeave Q1 2026 Earnings Transcript reported this week. LongYield Substack said the weighted average contract length is about five years. That gives the revenue stream more shape than the usual AI trade, where investors are often asked to trust vibes and a logo.

Customer quality is improving too, even if the base is still concentrated. CoreWeave said 10 customers are now committed to spend at least $1 billion each, including Anthropic and Meta, and it highlighted the $21 billion Meta agreement announced in April, CoreWeave Q1 2026 Earnings Transcript reported this week. That does not erase concentration risk. It does show that the customer list is broader than it was.

Then there is Nvidia, which remains a major strategic asset. CoreWeave said it raised $2 billion of equity in connection with the expansion of its relationship with Nvidia, CoreWeave Q1 2026 Earnings Transcript reported this week. my.stock.research describes Nvidia as a preferred partner with first allocation on new GPU generations. That is the outside author’s interpretation, not CoreWeave’s filing, but the broad point holds. Access to scarce top-end GPUs is part of the moat here.

The bear case begins with the obvious: CoreWeave can grow fast and still run into trouble if the capital structure gets ahead of the business.

CoreWeave said it has no material debt maturities until 2029, other than self-amortizing contract-backed debt and OEM vendor financing, CoreWeave Q1 2026 Earnings Transcript reported this week. That removes the classic near-term refinancing cliff. It does not make the debt load disappear. With interest expense already at $536 million in Q1 and headed higher in Q2, the company has little room for execution slippage.

Customer concentration remains another real issue. LongYield Substack reported that Microsoft accounted for about 67% of 2025 revenue, while OpenAI represented roughly 36% of FY2024 revenue. New contracts with Meta and Anthropic help at the margins, but this is still a narrow customer base for a company that depends on enormous, long-duration commitments. If one anchor customer changes terms, the damage would not be subtle.

Governance is the final layer, and it matters more than many bulls want to admit. CoreWeave disclosed material weaknesses in internal controls in its S-1, and said remediation efforts were expected to continue through 2025 and 2026, SEC.gov reported in March 2025. my.stock.research also noted the active federal securities class action, Masaitis v. CoreWeave, which alleges misleading statements about customer demand and reliance on a single third-party data center supplier. The case is unresolved, but unresolved is exactly what investors are paying for here.

One more thing deserves a blunt sentence. CoreWeave says the margin pressure is timing-based, not economic, and management expects adjusted operating margin to expand sequentially through the rest of the year and reach low double digits by Q4, CoreWeave Q1 2026 Earnings Transcript reported this week. If that happens, the bear case loses force. If it does not, the market will not keep giving the benefit of the doubt forever.

This is where the stock gets tricky. Cheap and expensive both miss the point.

my.stock.research laid out a probability-weighted model in April that put the implied value around $124 versus a current price of about $103, while assigning a 30% bear case of $25 to $45 per share. That model came before Q1 earnings, so the base case may have improved after CoreWeave beat earlier revenue expectations. The spread between outcomes is still wide enough to matter. This is not a tidy valuation call. It is a range of possible futures, some of them very attractive and some of them ugly.

That makes Q2 the next real test. CoreWeave guided Q2 revenue to $2.45 billion to $2.6 billion and adjusted operating income to $30 million to $90 million, CoreWeave Q1 2026 Earnings Transcript reported this week. If results land near the top of that range and the margin recovery starts to show through, the selloff begins to look like a market that overreacted to the pain and underweighted the payoff. If margins disappoint again, the discount starts to look earned.

That is why this is a speculative buy, not a clean one. CoreWeave looks appealing for investors who can tolerate balance-sheet strain and still want exposure to one of the fastest-growing infrastructure businesses in the market. It does not look suited to anyone who needs the numbers to behave before getting involved. The stock is asking for faith in a company that is still building the machine it needs to justify the valuation.

CoreWeave is one of those companies where the bull case and the bear case can both be true at the same time. The backlog is real, the customer demand is real, and CoreWeave Investor Relations said this week that Q1 was the strongest bookings quarter in the company’s history. The use is real too. So are the legal and governance overhangs.

That is why the answer to should you buy CoreWeave stock after the dip is conditional, not absolute. The stock may still have room to run, but only if the next two quarters confirm that CoreWeave can keep turning backlog into operating profit fast enough to outrun the debt and the drag that comes with it. If that happens, the dip looks like an opening. If it does not, it looks like a warning the market was early to spot.