MSFTMicrosoftNeutralMicrosoftIRENIREN LIMITEDBullishIREN LIMITED

3d ago on

I analyzed $IREN's conference call regarding its Q1 FY2027. Here are the key takeaways that are important for investors. Wall Street is selling off the Neocloud sector due to fears over capital expenditures, but IREN proves that this is not a "cloud-based WeWork," but rather a vertically integrated infrastructure monopoly. The gap between market sentiment and the company’s actual business performance has reached an extreme. Therefore, now is the best time to buy. Key management points that the market is overlooking: Absolute pricing power: IREN has just renewed its old H100 clusters for a multi-year term at a price twice as high as the original. New 5-year contracts are being signed at $22M/MW (short-term contracts exceed $40M/MW), with customers prepaying 90% of GPU CapEx. This sets a strong benchmark for returns on the scarce Vera Rubin chips. Immunity to debt risks: Unlike competitors with 20-year data center lease commitments, IREN owns the land, 5 GW of power capacity, and $6B (rising to $10B by June) worth of data centers, built with cash and no debt. Two-thirds of its backlog, totaling over $20B, comes from giants like $MSFT. The Microsoft Factor and the $4 B ARR Shift: The target date shift to January–February 2027 was driven by a request from MSFT itself to upgrade its network to support the latest frontier models. At the same time, Horizon 1 was delivered in 14 months, compared to the industry standard of 18 months. Capital discipline: Adjusting the remaining CapEx to $15B is a move made from a position of strength. The $6B requirement is covered by $12–15B in non-dilutive financing secured against net assets. My verdict: With an AI cloud gross margin of 86% and a GPU payback period of ~18 months, the market is mistakenly punishing IREN on par with over-leveraged intermediaries.

MSFTMicrosoftNeutralMicrosoftIRENIREN LIMITEDBullishIREN LIMITED

5d ago on

Big tech prepays 90% of GPU capex; $20.3B backlog sets up IREN revaluation despite possible local dip

Q1 FY27 Report for $IREN. Should You Buy or Not? This quarter marks a historic transformation: - AI Cloud revenue nearly tripled to $200M (up 182% QoQ) with a gross margin of 86%. The mining segment’s share fell to 12%. - Adjusted EBITDA soared from $19M to $266M. - The backlog reached $20.3B (+22% QoQ) with two-thirds coming from enterprise-level clients, led by $MSFT. The key insight is a shift in the economics of the deals. Customers are prepaying 90% of GPU capital expenditures under 5-year contracts ($22M/MW) and paying twice as much for H100s as before. By controlling >5 GW of scarce energy, IREN is shifting capital risk onto the buyers themselves. Moving the target for reaching $4+ billion in ARR from the end of the year to January–February could trigger a local sell-off. But when your CapEx is financed by the big tech companies themselves, a one-month delay is just noise ahead of a fundamental revaluation.

MSFTMicrosoftBullishMicrosoftAMZNAmazonBullishAmazonGOOGLAlphabetBullishAlphabetCRDOCredo Technology Group HoldingNeutralCredo Technology Group HoldingIRENIREN LIMITEDNeutralIREN LIMITEDMETANeutralNBISNebius Group N.V.NeutralNebius Group N.V.

6d ago on

What is causing the decline in AI favorites ($NBIS, $IREN, $CRDO)? And why are hyperscalers ($GOOGL, $AMZN, $MSFT) surging? The AI market paradox following the election stems from the possible end of the era of unchecked CapEx growth. A Democratic victory will trigger strict regulatory scrutiny. Add to this pressure on cash flows (S&P’s downgrade of its outlook for $META to negative) and a slowdown in AI spending growth by 2028: down to 16% for hyperscalers and 6% for niche cloud providers. For hyperscalers, the forced slowdown in CapEx is a blessing: the market is pricing in protection for FCF and credit ratings. But the giants’ CapEx is contractors’ revenue. Selective debt financing and regional barriers are weighing on the multiples of CRDO (-10%), NBIS (-7%), and IREN (-11%). My conclusion: the phase of expanding AI capacity at any cost is in danger of coming to an end. Because of this, capital is now shifting away from "shovels and picks" toward balance sheet strength and FCF yields. Therefore, it’s worth buying into the dip in AI infrastructure by betting on the strongest players.