This Artificial Intelligence (AI) Infrastructure Stock Has Been Crushing Nvidia on the Market. Its Latest Deal Could Send the Stock Even Higher

Applied Digital reported an estimated fiscal 2026 revenue increase of 96% to $422 million and disclosed a new 15-year, $5.2 billion lease with a U.S. hyperscaler for 210 MW of AI data-center capacity. Its lifetime lease revenue pipeline stands at $36 billion in the base case and could rise to $86 billion if renewal options are exercised, supporting the stock's premium valuation at 35 times sales. The article is broadly bullish on the company's long-term growth trajectory, though it is framed as investment commentary rather than new company guidance.

Analysis

APLD is transitioning from a narrative stock to a duration asset: the market is starting to underwrite multi-year contracted cash flows rather than spot revenue. That changes the competitive field because the real scarce input is not GPU demand, but power, land, interconnect, and permitting — all of which create a higher barrier to entry and make incumbent hyperscaler-backed capacity far harder to displace. The second-order winner is the ecosystem of electrical equipment, switchgear, transformers, and utility-facing contractors; the loser set is any smaller colo player without the balance-sheet depth to pre-secure megawatts.

The key risk is that the equity is now being valued like an option on flawless execution, while the buildout path is still lumpy and capital intensive. The gap between signed lease value and near-term free cash flow can stay wide for multiple quarters if construction timelines slip, interconnect queues extend, or financing terms tighten. In that scenario, the stock can de-rate hard even with intact demand, because the market tends to punish data-center names when the story shifts from "contracts signed" to "cash conversion delayed."

Consensus is probably underestimating how much of the upside is already in the multiple, but also underestimating how sticky customer demand could be if APLD becomes a critical capacity bottleneck. The best setup is not chasing strength blindly, but expressing a view that the business de-risks over time while the stock remains highly sensitive to each new lease, financing update, and capacity milestone. NVDA is an indirect beneficiary only insofar as APLD expands the AI deployment funnel; the real trade is infrastructure scarcity, not chip volume.

Key Decisions for Investors

• Stay long APLD only as a catalyst-driven position, not a core holding; size small and use a 1-3 month horizon into the next lease/financing update, because the stock can re-rate 20-30% on execution but also give back the move if capex timing slips.
• Pair trade: long APLD / short a lower-quality AI infrastructure or colocation peer with weaker contracted visibility and tighter financing flexibility; the spread should work if investors continue rewarding backlog quality over headline growth.
• For existing NVDA exposure, consider it a second-order beneficiary of AI capex expansion rather than a direct hedge; add on APLD pullbacks only if you want broader AI infrastructure beta without paying APLD's higher execution risk.
• Use call spreads instead of outright stock in APLD if initiating fresh exposure: a 3-6 month bull call spread captures upside from new contract announcements while limiting drawdown from valuation compression.
• If APLD trades up on another contract headline, trim into strength rather than chase; the risk/reward worsens materially once the market fully capitalizes additional lease value before any corresponding cash conversion.