Why IREN Limited Rallied Almost 40% in May

IREN reported Q3 revenue of $144.8 million, down 21.6%, and adjusted EBITDA of $59.5 million, down 20% quarter over quarter, as it continues shifting away from Bitcoin mining toward AI cloud services. The company announced two acquisitions—Mirantis for $625 million in stock and Nostrum Group in Spain—plus a partnership with Nvidia to deploy up to 5 GW of DSX architecture, supported by a five-year $3.4 billion cloud contract and a warrant for 30 million shares at $70. IREN also raised nearly $3 billion in convertible notes, improving funding for its build-out despite ongoing execution and capital intensity risks.

Analysis

The market is starting to price IREN less like a cyclical miner and more like an option on scarce AI power. The real second-order effect is that Nvidia’s economic commitment can re-rate IREN’s land/power portfolio ahead of near-term earnings, because the value of gigawatts with interconnects, not current EBITDA, becomes the key scarce asset. That helps IREN’s capital raise math and may compress financing costs further if the market believes the NVIDIA relationship reduces offtake risk.

The acquisition mix is strategically coherent: Mirantis pushes IREN up the software stack, while Spain gives it a credible non-U.S. expansion path where permitting and power economics can be more favorable. The competitive implication is that smaller neoclouds without balance-sheet depth or utility-scale power access may get squeezed into niche inference workloads, while hyperscalers face a new vertically integrated rival with a lower cost of power and more flexible site control. This is especially relevant if AI factory deployments increasingly value dedicated capacity over generic cloud elasticity.

The biggest risk is not execution rhetoric; it is timeline mismatch. IREN is effectively pre-monetizing multi-year capacity with near-term debt and equity-like instruments, so any delay in data center buildout, GPU availability, power delivery, or customer demand could force more dilutive capital raises before the Sweetwater asset is cash generative. A second-order bear case is that Nvidia’s involvement is as much strategic control as endorsement, which could cap IREN’s negotiating leverage and keep economics concentrated upstream with the chip vendor.

The consensus seems to underappreciate how much of the upside is already being pulled forward by narrative and warrant structure. If the stock stalls below the warrant strike, IREN may still be forced to fund growth at less attractive terms, creating a valuation overhang despite positive headlines. The asymmetry is that the equity can continue to rip on partnership headlines, but the operating story only validates if construction milestones and contracted AI revenue scale meaningfully over the next 6-18 months.

Key Decisions for Investors

• Long IREN on pullbacks over the next 1-3 weeks, but size as a financing-driven momentum trade rather than a fundamental core holding; upside remains substantial if the Nvidia relationship is interpreted as strategic validation, but the position should be monitored for dilution risk.
• Buy IREN call spreads 3-6 months out rather than outright stock to capture continued re-rating from additional AI/power announcements while defining downside if the market refocuses on funding and execution risk.
• Pair trade: long IREN / short a basket of smaller AI-neocloud or high-cost computing names over 1-3 months, expressing the view that scale, power access, and Nvidia adjacency will concentrate capital toward the strongest platform.
• For more risk-averse exposure to the theme, prefer NVDA over IREN on any spread-widening caused by capital intensity concerns; Nvidia monetizes the AI infrastructure buildout with materially lower balance-sheet risk.
• Take profits on IREN into strength if the stock approaches levels where future financing becomes easier but incremental upside from warrants and headlines diminishes; the trade is likely to become crowded quickly after any new catalyst.