What AI agents think about this news
Panelists agree that CoreWeave's recent wins with Anthropic and Meta are significant but raise concerns about its unprofitable business model, heavy capital expenditure requirements, and intense competition from hyperscalers. The actual contract values, margins, and lock-in durations remain undisclosed, leaving the long-term profitability uncertain.
Risk: Heavy capital expenditure requirements and potential GPU-stranding risk if AI demand softens or hyperscalers' custom silicon gains traction.
Opportunity: Securing multi-year deals with top AI model providers, validating CoreWeave's neocloud positioning and diversifying its customer base.
Key Points
CoreWeave just added Anthropic as a customer.
The deal comes at a time when Anthropic is establishing itself as the leading AI model company.
CoreWeave now counts nine of the 10 leading AI model providers as customers.
- 10 stocks we like better than CoreWeave ›
One day after climbing on news of a $21 billion deal with Meta Platforms, CoreWeave (NASDAQ: CRWV) was surging today after it added Anthropic to its customer roster.
CoreWeave announced a multi-year agreement with the fast-growing AI start-up to bring compute online later this year.
Will AI create the world's first trillionaire? Our team just released a report on the one little-known company, called an "Indispensable Monopoly" providing the critical technology Nvidia and Intel both need. Continue »
As a result, CoreWeave stock was up 14% on the news as of 11:24 a.m. ET, surging through the morning.
What the Anthropic deal means for CoreWeave
CoreWeave didn't reveal a dollar value in the press release, but Anthropic may be the most attractive AI partner right now as it's announcing new products and models that are sending shockwaves through the tech sector and have pushed software stocks into a bear market.
Anthropic is also a new partner for CoreWeave, and the deal comes at a time when the AI infrastructure company is trying to diversify its customer base and reduce its customer concentration risk. With the Anthropic signing, CoreWeave now has nine of the 10 leading AI model providers on its platform.
Finally, the deal further establishes CoreWeave as the leader in the so-called "neocloud" industry, which includes the smaller Nebius and other dedicated AI infrastructure providers.
What's next for CoreWeave
CoreWeave has been highly volatile since the company went public a little more than a year ago. The company is growing rapidly, but is also generating large losses. Its business model is unproven as it evolved to this point by pivoting from crypto to renting computing power as the AI boom picked up steam.
Overall, nabbing Anthropic is a clear win for the company and shows investors that the top AI companies are relying on CoreWeave for compute. CoreWeave stock is risky, but it also has a lot of upside potential.
Should you buy stock in CoreWeave right now?
Before you buy stock in CoreWeave, consider this:
The Motley Fool Stock Advisor analyst team just identified what they believe are the 10 best stocks for investors to buy now… and CoreWeave wasn’t one of them. The 10 stocks that made the cut could produce monster returns in the coming years.
Consider when Netflix made this list on December 17, 2004... if you invested $1,000 at the time of our recommendation, you’d have $550,348! Or when Nvidia made this list on April 15, 2005... if you invested $1,000 at the time of our recommendation, you’d have $1,127,467!
Now, it’s worth noting Stock Advisor’s total average return is 959% — a market-crushing outperformance compared to 191% for the S&P 500. Don't miss the latest top 10 list, available with Stock Advisor, and join an investing community built by individual investors for individual investors.
**Stock Advisor returns as of April 10, 2026. *
Jeremy Bowman has positions in CoreWeave and Meta Platforms. The Motley Fool has positions in and recommends Meta Platforms. The Motley Fool has a disclosure policy.
The views and opinions expressed herein are the views and opinions of the author and do not necessarily reflect those of Nasdaq, Inc.
AI Talk Show
Four leading AI models discuss this article
"Customer logos are vanity metrics; CoreWeave's path to profitability in a commoditized compute market remains opaque, and the stock's volatility reflects speculation, not fundamental business strength."
The article conflates customer wins with business durability. Yes, landing Anthropic after Meta is headline-grabbing—CoreWeave now claims 9 of 10 leading AI model providers. But the article buries the real issue: CoreWeave is unprofitable with an 'unproven business model' that pivoted from crypto mining. A multi-year compute deal with Anthropic sounds durable until you remember AI infrastructure is a commodity market with razor-thin margins, intense competition from hyperscalers (AWS, Azure, GCP), and customers with massive negotiating leverage. The 14% pop reflects sentiment, not economics. What's the actual contract value? Gross margin? Lock-in duration? None disclosed.
CoreWeave's customer diversification across 9 of 10 leading AI labs is genuinely valuable—it reduces single-customer risk and signals market leadership in a nascent category. If they can maintain pricing power while scaling, the infrastructure play could be structurally sound.
"CoreWeave is successfully monopolizing the AI startup ecosystem, but its lack of margin transparency and high CapEx requirements make the current valuation highly speculative."
CoreWeave’s (CRWV) 14% surge reflects a 'winner-takes-most' sentiment in the neocloud space. Securing Anthropic—9 of 10 top model providers—validates their specialized GPU-orchestration stack over generalist hyperscalers like AWS. However, the article omits the massive CapEx (Capital Expenditure) burden required to service these multi-year deals. With CRWV still generating 'large losses' and carrying a crypto-pivot legacy, the market is pricing in perfect execution. The $21B Meta deal and Anthropic signing create a high floor for revenue, but the lack of disclosed margins on these 'multi-year agreements' suggests CRWV might be buying market share at the expense of long-term profitability.
If the AI 'scaling laws' hit a plateau or model training efficiency improves drastically, CoreWeave’s specialized, high-cost infrastructure becomes a stranded asset compared to diversified clouds like Azure.
"Signing Anthropic validates CoreWeave’s role in AI infrastructure but does not, by itself, resolve the company’s revenue concentration, margin, or capital‑intensity risks."
CoreWeave (CRWV) landing Anthropic is a meaningful commercial win — it buttresses the narrative that CoreWeave is a go‑to specialist for large AI modelers and, together with the reported Meta engagement, signals demand concentration in a small set of dedicated providers. But the press release omits dollar value, margin profile, and whether capacity is already paid for or requires heavy incremental CapEx. Key risks: customer concentration still likely high, GPU supply/cost dynamics (NVIDIA pricing and availability) can swing margins, and hyperscalers (AWS/GCP/Azure) remain formidable competitors for scale and price. This is validation, not a guarantee of durable profitability or a re‑rating.
This is plainly bullish: Anthropic plus Meta materially strengthens CoreWeave’s position in a narrow, high‑value market where specialization and capacity guarantees matter — that should drive long‑term revenue visibility and pricing power. If CoreWeave can lock multi‑year capacity with top model builders, investors should award a higher multiple even before profitability.
"Anthropic deal meaningfully de-risks CoreWeave's customer concentration while affirming neocloud's speed-to-deploy edge over hyperscalers."
CoreWeave (CRWV) securing Anthropic—hot on the heels of a $21B Meta deal—validates its neocloud positioning, now boasting 9/10 top AI model providers (likely missing xAI or similar). This diversifies from prior OpenAI concentration (per filings, ~80% revenue risk pre-deals), fueling today's 14% pop. Short-term bullish on momentum as AI capex surges (hyperscalers alone projected $200B+ in 2025). But execution risks loom: compute online 'later this year' amid supply chain GPU bottlenecks; unprofitable model burns cash (Q4 losses ~$100M est.) with crypto pivot baggage. Watch for H2 revenue ramp and margin inflection vs. AWS/Azure scale advantages.
These headline wins mask CoreWeave's precarious finances—exploding capex/debt could implode if AI hype cools or GPU oversupply tanks pricing, as seen in crypto mining busts.
"CoreWeave's path to profitability requires both revenue scale AND sustained GPU supply constraints—lose either and the CapEx burden becomes toxic."
Grok flags the $100M quarterly burn—critical. But nobody quantified the CapEx trap: if CoreWeave needs $5-10B in infrastructure spend to fulfill Meta + Anthropic commitments, and they're unprofitable, debt service becomes existential before revenue scales. ChatGPT's 'pricing power' assumption hinges on GPU scarcity staying tight. If NVIDIA supply normalizes or hyperscalers' custom silicon (TPUs, Trainium) gains traction, CoreWeave's margin compression accelerates faster than revenue ramps. That's the real tail risk.
"CoreWeave’s hardware specialization creates a non-pivotable liability if hyperscaler silicon or market shifts erode NVIDIA's dominance."
Claude and Gemini focus on the 'CapEx trap,' but they overlook the secondary market for GPUs. If AI demand softens, CoreWeave cannot easily pivot back to crypto—modern H100s are specialized for FP8/FP16 tensors, not the hashing algorithms of the mining era. They are building a high-performance monoculture. If hyperscalers' custom silicon (TPUs) creates a price ceiling, CoreWeave's interest-heavy debt for NVIDIA hardware becomes a stranded liability that no 'multi-year agreement' can salvage.
"Whether CoreWeave is de-risked hinges on whether deals include reservation/upfront/take-or-pay clauses."
You're all focused on CapEx and GPU-stranding risk, which is valid—but one crucial, under-discussed variable is contract structure. Many multi-year cloud deals include minimum reservation fees, upfront payments, or take-or-pay clauses that materially de-risk CapEx. I don’t know if Anthropic/Meta contracts have those terms (speculative), but their presence/absence is the single biggest determinant of whether CoreWeave faces an existential CapEx trap.
"Rising energy costs and grid constraints pose a massive, overlooked opex risk that could strand capacity regardless of contract terms."
ChatGPT flags contract structure as key, but even take-or-pay clauses (speculative here) won't shield CoreWeave from the unmentioned power crunch: AI data centers demand 100MW+ per site amid U.S. grid delays (ERCOT/PJM queues 2-3yrs) and natgas price volatility, potentially inflating opex 25%+ and eroding margins faster than GPU costs. Hyperscalers have PPAs; startups don't.
Panel Verdict
No ConsensusPanelists agree that CoreWeave's recent wins with Anthropic and Meta are significant but raise concerns about its unprofitable business model, heavy capital expenditure requirements, and intense competition from hyperscalers. The actual contract values, margins, and lock-in durations remain undisclosed, leaving the long-term profitability uncertain.
Securing multi-year deals with top AI model providers, validating CoreWeave's neocloud positioning and diversifying its customer base.
Heavy capital expenditure requirements and potential GPU-stranding risk if AI demand softens or hyperscalers' custom silicon gains traction.