Is Iren Stock a Buy Now?

Iren is exiting Bitcoin mining by year-end and pivoting to AI cloud infrastructure, with AI cloud revenue rising nearly 685% to $128.8 million in fiscal 2026 from $16.4 million. The company secured a $9.7 billion, five-year Microsoft cloud-services contract expected to generate roughly $1.94 billion of annualized revenue at full deployment, but posted a $702.6 million net loss after a $638.8 million mining-hardware impairment. Iren plans $25 billion to $30 billion of 2027 capex for AI data-center expansion, creating substantial execution, funding, and leverage risk despite strong AI compute demand.

Analysis

IREN is moving from a volatile, self-funded commodity business into a contracted-infrastructure model, but the valuation hinge is not AI revenue growth; it is the spread between contracted GPU rents, power costs, hardware depreciation, and a rapidly rising cost of capital. The planned build implies an asset base far larger than the current operating history can support, so equity holders are effectively long both hyperscaler demand and continued availability of project finance. Customer prepayments reduce initial funding needs but do not eliminate residual capex, refinancing, or GPU-obsolescence risk.

The more investable second-order beneficiary is NVDA: an accelerated GPU deployment cycle supports data-center demand regardless of whether IREN earns attractive returns on that hardware. MSFT gains incremental capacity without owning all the physical infrastructure, but concentration creates an asymmetry for IREN: any commissioning delay, utilization dispute, or counterparty reprioritization would have an outsized effect on its equity and debt terms. Competing GPU-cloud providers such as CRWV and NBIS may face lower pricing power if IREN’s capacity arrives on schedule, particularly for frontier-training clusters.

Over the next 1-3 months, the key catalyst is evidence that financing is committed at tolerable rates and that commissioned capacity converts to recognized revenue rather than merely contracted backlog. Over 6-18 months, the structural risk is that GPU generations turn over faster than contract economics amortize, compressing renewal pricing and leaving operators with stranded or under-earning power infrastructure. The contrarian view is that the selloff may be insufficient if the market is capitalizing headline contract value without applying a project-finance discount to execution risk; conversely, credible non-recourse financing and phased commissioning would justify a sharp multiple rerating.

Key Decisions for Investors

• Keep IREN on a financing/commissioning watch rather than initiate a core long before capex funding is independently disclosed. Upgrade only after committed debt/equity sources cover the next construction phase and management demonstrates on-time revenue recognition; falsify the constructive case if funding requires repeated dilutive equity issuance or contracted run-rate is pushed out by more than one quarter.
• Express AI-infrastructure demand through a long NVDA / short IREN pair over 3-6 months for investors seeking lower execution risk: NVDA captures GPU shipments while IREN bears construction, customer-concentration, and balance-sheet risk. Reassess if IREN secures largely non-recourse financing with fixed-rate funding or if NVDA demand indicators weaken materially.
• Avoid treating IREN as a clean crypto-beta substitute during the transition. If residual mining exposure or power assets retain meaningful value, Bitcoin upside can mask deteriorating cloud unit economics; monitor segment gross margin, GPU utilization, and cash interest coverage rather than headline cloud backlog.
• For a tactical IREN long, wait for a post-financing selloff rather than chase contract announcements; size as a high-volatility special situation with a 6-12 month horizon. Target upside requires delivered capacity and lower perceived dilution, while downside is substantial if project funding costs rise or customer commitments are delayed.