BullOrBSBullOrBS
7/10
PICKNBIS3 min read

Four AI Infrastructure Giants: Which Bet on GPU Cloud Makes Sense in 2026?

We're comparing four companies racing to build AI data centers—APLD, CRWV, NBIS, and IREN. All are unprofitable or early-stage, all depend on Nvidia, and all are extremely volatile. NBIS just got a $2B Nvidia vote of confidence and has the wildest growth story. CRWV is the biggest revenue-wise but drowning in debt. IREN is the only profitable one but pivoting away from crypto. APLD is the smallest and riskiest. Winner: NBIS—if you can stomach the volatility.

Data sourced March 2026. Verify current figures before making investment decisions.

The Verdict

AI EDITORIAL OPINION

Score: 7/10 — NBIS is worth a serious look if you can handle volatility.

NBIS is the growth leader here: $1.25B ARR in 2025 (beat guidance) pointing to $7–9B in 2026. Nvidia just invested $2B, which is not a casual endorsement—it's a structural bet. Analysts rate it Strong Buy with 36% upside to $150+. But here's the honest truth: this is a high-risk, high-reward play. There's no dividend, the company doesn't make money yet, and execution risk is real. If Nebius delivers 2026 guidance and Nvidia keeps funding it, you could see 50%+ returns. If capex slows or the company stumbles on execution, you could lose 30–50%.

If you're putting this in a TFSA and can afford to ride swings, NBIS beats APLD (too volatile, too small), CRWV (too much debt, new IPO uncertainty), and IREN (stuck in a crypto pivot). Would I buy? Yes—but only with money I can afford to lose, and with a 2–3 year time horizon. Don't buy on margin, and don't buy the dip every time it crashes. Position size accordingly.

Disclaimer

This analysis is AI-generated by BullOrBS for educational and entertainment purposes only. It is not financial advice. BullOrBS is not affiliated with any financial publication, newsletter, or institution mentioned in our analysis. Always do your own research and consult a qualified financial advisor before making investment decisions.

Every stock we evaluated, and why most didn't make the cut:

CutAPLDApplied Digital Corporation
4/10

Smallest revenue base, highest beta (7.34), customer concentration risk after Nvidia exit, and currently down 36% from highs despite analyst optimism.

CutCRWVCoreWeave, Inc.
5/10

Lost $1.2B in net income TTM despite high revenue, down 59% from $187 peak, and faces execution risk as a brand-new public company with enormous leverage.

ReviewedIRENIREN Limited (formerly Iris Energy)
6/10

Only profitable stock in this group with P/E of 44x and $688.6M TTM revenue. Revenue growth of 355% YoY and $9.7B Microsoft AI Cloud contract provides real revenue visibility. Analyst target of $79.78 implies 106% upside. Lowest beta risk at 1.80–2.84 compared to peers.

Head-to-Head

MetricAPLDCRWVNBISIREN
Market Cap$7.7B$39.1B~$25B$12.7B
Revenue (TTM/Recent)~$264M$5.1B$529M revenue (2025) / $1.25B ARR$688.6M
Profitable?No (−$0.39 EPS)No (−$2.81 EPS, −$1.2B net loss)No (early-stage ramp)Yes (44x P/E)
Revenue Growth Rate~250% YoY (Q2 FY2026; TTM ~83%)~134% YoY~547% YoY (Q4)~355% YoY (Q1)
Volatility (Beta)7.34N/A (new IPO)N/A (relisted 2024)1.80–2.84
Analyst Target$45.27 (+61%)$124.70 (+63%)$150.62 (+36%)$79.78 (+106%)
Key StrengthAnalyst consensusRevenue scale + marginsNvidia partnership + hypergrowthProfitability + diversity
Key RiskCustomer concentration, leverage, volatilityMassive debt, new IPOYandex heritage, executionBTC exposure, pivot execution

What each number means:

  • Revenue Growth: Shows how fast each company is expanding. NBIS is growing nearly 4x faster than CRWV, even though CRWV has 10x more revenue. This is the hockey-stick moment.
  • P/E Ratio (Profit Multiple): Only IREN has one because it's profitable. The others lose money, so you're betting purely on future profits. Higher P/E = market expects more growth (IREN at 44x is expensive for a mature company, cheap if it's growing fast).
  • Beta (Volatility): APLD's 7.34 means if the S&P 500 drops 10%, APLD could drop 70%. IREN's 1.80–2.84 means it swings 2–3x the market—still high, but manageable.

Why NBIS Wins

NBIS just got a $2 billion vote of confidence from Nvidia TODAY. Think of it this way: Nvidia is like a manufacturer of ovens. Nebius is building bakeries. Nvidia just wrote a $2B check to help Nebius build faster. That's not a casual investment—it's a strategic bet that Nebius will be a major customer of Nvidia GPUs for years to come.

The numbers back it up. Nebius's 2025 ARR (annualized revenue run rate) hit $1.25 billion—it beat its own target by $250 million. Now it's guiding $7–9 billion in ARR for 2026. That's roughly a 6–7x jump. Yes, it's a projection. But the company is executing faster than expected, and Nvidia is putting real capital behind it.

Compare that to APLD (smallest, losing money, highest volatility) or CRWV (biggest revenue but drowning in debt and down 59% from its peak). IREN is the safest choice because it's profitable, but it's still mostly a Bitcoin mining company pivoting to AI—the execution risk is real. NBIS is the company where the best-in-class supplier (Nvidia) is literally financing growth. That's a rare signal.

The Runner-Up

IREN is the most defensible alternative—it's the only profitable company here, has real revenue contracts ($9.7B Microsoft deal), and lower volatility. The 106% analyst upside target is compelling. But here's the catch: over 90% of IREN's revenue still comes from non-AI operations (BTC mining + data center hosting), which just crashed. The AI cloud business is growing fast (137% QoQ) but is still under 10% of revenue. It's a pivot play, not a pure AI growth story. NBIS, by contrast, is the pivot—it started as Yandex, restructured, and is now a pure-play AI cloud company. Less legacy baggage, clearer thesis.

NBIS Current Price

$111.71 (up 16% today after Nvidia news)

Robinhood (March 11, 2026 intraday)

NBIS 2025 ARR (Actual)

$1.25B (beat target by $250M)

StockAnalysis

NBIS 2026 ARR Guidance

$7–9B (6–7x growth)

StockAnalysis

NBIS Analyst Consensus

Strong Buy (8 analysts), Avg Target $150.62–$151.71

Public.com / StockAnalysis

NBIS Nvidia Partnership

$2B investment announced March 11, 2026 for AI data center buildout

Robinhood / Company news

APLD Beta (Volatility)

7.34 (swings 7x harder than the market)

Yahoo Finance

CRWV Revenue (TTM)

$5.1B with −$1.2B net loss

StockTitan

IREN Revenue (TTM)

$688.6M with P/E of 44x (only profitable stock)

TickerNerd / Robinhood

IREN Microsoft AI Cloud Contract

$9.7B committed over multi-year term

StockAnalysis

Risks They Missed

  • All four depend on Nvidia GPU supply and continued AI spending growth—if capex slows or Nvidia reallocates chips, all four crater simultaneously.
  • NBIS carries geopolitical risk from its Yandex heritage (Russia ties, potential sanctions/regulatory scrutiny even post-restructuring).
  • CRWV and NBIS carry enormous debt from GPU financing; if interest rates stay high or revenue slows, both face refinancing pressure.
  • APLD has customer concentration risk—after Nvidia and CoreWeave issues, the company is vulnerable if a major customer pulls back.
  • IREN's non-AI exposure (over 90% of current revenue from BTC mining + data center hosting) makes it a hybrid play that could underperform if crypto crashes while AI ramps.
  • None of these companies are profitable on a GAAP basis—you're betting on future earnings, not current cash generation.

Catalysts

  • NBIS: Nvidia's $2B investment closes (likely Q2–Q3 2026) and company executes 2026 revenue guidance ($4B)—could trigger 20–50% upside if margins hold.
  • CRWV: Company reduces debt or reaches free-cash-flow positive status—would validate the investment thesis and reduce refinancing risk.
  • IREN: AI cloud business grows to 30%+ of revenue and achieves profitability in AI segment—would prove the pivot is real, not just a rebranding.
  • APLD: Secures a major long-term customer contract (not Nvidia or CoreWeave) to reduce concentration risk and stabilize the stock.
  • Industry-wide: Nvidia announces new GPU allocation or pricing changes—could either accelerate capex (bullish) or trigger margin compression (bearish) for all four.

NEXT ANALYSIS

PLTR vs. BBAI: David vs. Goliath in the AI-Defense Arms Race

Want more analysis like this?

Get AI-driven stock analysis in your inbox every week. Free.

By subscribing, you agree to our Privacy Policy and consent to receiving emails from BullOrBS. Unsubscribe anytime.