What Is One of the Best AI Stocks to Buy Right Now?

IREN is targeting $3.4 billion in annualized revenue by the end of 2026, but that would use only about 10% of its secured power capacity, implying a potential long-term scale to roughly $30 billion or more. The company has already secured a $9.7 billion five-year Microsoft deal and says it is in advanced negotiations with additional hyperscalers, while hiring in Australia signals further data center expansion. The article frames IREN as a leveraged beneficiary of the AI data center spending boom, though much of the upside remains speculative.

Analysis

IREN is becoming a scarcity asset rather than just a growth story: the market is starting to price a portfolio of power-constrained infrastructure with embedded optionality on multiple hyperscaler customers. The second-order implication is that incremental wins should compress the perceived customer-concentration discount and expand valuation multiples faster than revenue changes alone, because each additional contract reduces perceived execution risk and increases the monetization rate of already-secured capacity.

The bigger read-through is competitive: air-cooled, faster-deploying capacity is likely to win share versus more complex liquid-cooling builds over the next 6-18 months, especially for customers trying to bring inference online quickly. That favors operators with land, interconnects, and permitting already in hand, while pressuring smaller developers who need fresh capital, longer lead times, and more bespoke engineering. It also creates a subtle supply-chain benefit for electrical infrastructure vendors and grid-connected industrial service names, as buildouts shift from speculative to contracted deployment.

The main risk is that the market may be extrapolating too much of the capacity stack too early. If hyperscaler demand normalizes, or if financing/re-pricing forces a slower conversion of secured power into signed revenue, the equity can de-rate hard because the bull case is already predicated on high utilization and rapid contract closure. The time horizon matters: near-term catalyst is another contract announcement over the next 1-3 months; the failure mode is a 2-4 quarter gap between power availability and monetization, which would expose the stock’s multiple to mean reversion.

Consensus is likely underestimating how much of the upside is operational execution, not just AI demand. The market is treating IREN like a pure beneficiary of the AI capex cycle, but the real edge is balancing power access, build velocity, and commercial credibility; that combination can turn a single anchor customer into a multi-customer leasing platform. If management sustains that conversion rate, the stock can remain in a re-rating phase longer than the typical ‘hot AI name’ trade.

Key Decisions for Investors

• Long IREN into contract-announcement catalysts over the next 1-3 months; best risk/reward is if the next deal confirms multi-hyperscaler demand, with upside from multiple expansion rather than just revenue revisions.
• Buy IREN call spreads with 6-12 month expiry to express continued re-rating while limiting downside if monetization lags; prefer strikes that assume only partial conversion of secured power capacity.
• Pair trade: long IREN / short a basket of higher-leverage AI infrastructure developers without equivalent contracted power or customer validation; thesis is that credibility and time-to-revenue become the key differentiators.
• Add MSFT on any weakness as the lead customer validating the ecosystem, but size modestly; the better trade is the supplier/landlord side rather than the hyperscaler itself, where AI capex is already well-discounted.
• Set a tactical stop on IREN if no new customer wins emerge within 2 quarters; absence of follow-through would signal the market has over-advanced the multiple versus actual revenue conversion.