Applied Digital Corporation (NASDAQ: APLD) has signed a landmark 15-year lease with a U.S.-based investment-grade hyperscaler valued at approximately $5 billion, marking one of the largest AI infrastructure agreements announced by any publicly traded data-center developer this year. The deal covers 200 megawatts (MW) of capacity at the company’s upcoming Polaris Forge 2 Campus in North Dakota, a next-generation “AI factory” designed to support large-scale training, inference, and high-performance computing (HPC) workloads.
The agreement also includes a right of first refusal (ROFR) for an additional 800 MW, which, if exercised, could lift total contracted capacity to one gigawatt (GW). That expansion potential positions Polaris Forge 2 as one of the largest AI-dedicated campuses in the continental U.S.—and firmly situates Applied Digital among the few mid-cap players now competing with hyperscale incumbents like Amazon Web Services, Microsoft Azure, and Google Cloud for next-gen compute real estate.
How Applied Digital’s $5 billion lease reshapes hyperscaler infrastructure demand in the AI era
The structure of the lease reflects a seismic shift in hyperscaler infrastructure strategy. Instead of vertically integrating all AI compute facilities, major cloud players are increasingly outsourcing power-dense buildouts to specialist developers such as Applied Digital. In this case, the 200 MW commitment represents a turnkey leasing model where the hyperscaler secures both data-hall space and critical IT load, while Applied Digital retains ownership of the physical assets and long-term operational control.
The Polaris Forge 2 site itself spans roughly 900 acres, adjacent to the company’s operational Polaris Forge 1 Campus, bringing the total AI-factory capacity in North Dakota to 600 MW. Designed with power-use effectiveness (PUE) below 1.18 and near-zero water consumption, the facility embodies Applied Digital’s blueprint for “AI factories”—industrial-scale data hubs optimized for sustained 30 kW per rack density, direct-connect power infrastructure, and renewable-energy integration.
Construction of the first 200 MW phase will begin immediately, with initial activation targeted for 2026 and full energization by 2027. Company executives indicated that ongoing negotiations with state regulators and utility providers have ensured access to low-cost energy and favorable permitting schedules, strengthening North Dakota’s emergence as a competitive hub for AI workloads.
Why the North Dakota location matters for hyperscaler diversification and AI energy economics
North Dakota may seem far removed from traditional Silicon Valley data-center corridors, but the region offers compelling economics. With stable power prices, abundant wind-energy generation, and naturally cold ambient temperatures, operators can achieve dramatic efficiency gains. The state’s regulatory climate also allows for accelerated land-use approvals and multi-decade grid-tie contracts—advantages that coastal markets like Virginia and California struggle to replicate amid grid congestion and environmental scrutiny.
Applied Digital’s management has long advocated that AI compute capacity is becoming a distributed resource, shifting from cloud concentration toward geographically diversified industrial campuses. CEO Wes Cummins has previously emphasized that execution velocity, not demand, is the bottleneck in AI infrastructure delivery. The company’s ability to deploy hundreds of megawatts in high-availability environments has been a recurring differentiator in its recent earnings calls.
From an economic standpoint, the lease’s 15-year tenor provides predictable cash flows resembling a utility-like annuity. With projected aggregate revenue near $5 billion, Applied Digital’s backlog now exceeds the company’s total market capitalization by multiples, giving investors a clearer line of sight into long-term monetization of AI capacity rather than short-cycle crypto-hosting contracts that once defined the firm’s earlier business model.
How the lease transforms Applied Digital’s balance sheet and investor sentiment toward AI infrastructure equities
In market terms, Applied Digital’s announcement catalyzed an immediate sentiment shift. Following the press release, APLD shares rose more than 4 percent in pre-market trading and traded near the $29–$30 range by mid-session. Investors interpreted the news as a validation of the company’s pivot from speculative compute hosting to institutional-grade AI infrastructure.
As of late-day trading on October 22, the stock recorded an intraday high of $35.84 and closed around $29.41 on elevated volume exceeding 35 million shares—its heaviest since mid-summer. Analysts covering small-cap technology infrastructure noted that the deal’s long-term recurring revenue profile could materially de-risk cash-flow forecasts for fiscal 2026 and beyond.
Institutional flows mirrored that optimism. Quantitative tracking of 13F filings indicates growing exposure from AI-thematic exchange-traded funds and infrastructure-focused hedge funds, positioning Applied Digital within the same narrative arc as CoreWeave, DigitalBridge, and Equinix. The key differentiator remains Applied Digital’s willingness to build in secondary geographies—the “heartland hyperscale” thesis—that offer both regulatory stability and energy resilience.
While bullish sentiment dominates, the absence of a publicly named tenant introduces a degree of opacity. The hyperscaler is described only as “U.S.-based, investment-grade,” leaving open speculation that it could be one of the top three cloud providers or a defense-aligned AI contractor. Any eventual disclosure could further clarify the credit quality underpinning the $5 billion commitment and influence future valuation multiples.
What financial analysts and infrastructure experts will watch as the project advances through 2027
The key performance indicators analysts are monitoring include construction pacing, energy-efficiency metrics, and tenant-activation milestones. If the first 100 MW comes online on schedule in 2026, Applied Digital could recognize material lease revenue sooner than consensus models currently anticipate.
The projected PUE of 1.18 and near-zero water use, if validated by third-party audits, would position Polaris Forge 2 among the most efficient AI facilities globally. The company’s design leverages indirect-evaporative cooling and on-site power redundancy architecture to maintain 99.999 percent uptime while minimizing environmental footprint—an increasingly salient differentiator as hyperscalers face sustainability disclosure requirements.
Equity researchers are also weighing whether the ROFR for an additional 800 MW will be exercised. Should the tenant commit to the full gigawatt, Applied Digital’s recurring revenue base could increase several-fold, establishing a cumulative contract value above $25 billion across its campuses. Such scale would rival early-stage lease volumes once associated with Amazon’s IAD and Microsoft’s Des Moines campuses in the late 2010s.
How the Polaris Forge 2 deal influences broader AI-factory economics and sector investment flows
At a macro level, this transaction exemplifies how hyperscaler leasing is redefining data-center economics. Traditional colocation facilities averaged three- to five-year terms; AI factory leases now stretch past 15 years, effectively transforming infrastructure developers into quasi-utilities whose valuations reflect discounted cash-flow stability rather than cyclical hardware margins.
The broader implication is a deepening bifurcation between legacy data-center real-estate investment trusts (REITs) and AI-specialized developers like Applied Digital. While REITs such as Digital Realty Trust and CyrusOne focus on multi-tenant retail colocation, Applied Digital’s single-tenant AI model yields higher energy throughput per square foot, supporting premium lease rates.
Capital markets have responded by rerating the subsector. AI infrastructure equities now trade at earnings multiples once reserved for semiconductor designers, underscoring how the “picks-and-shovels” layer of AI—power, land, and cooling—is absorbing investor enthusiasm formerly concentrated on model developers and GPU suppliers.
Applied Digital’s management has described its AI factories as “industrial operating systems for intelligence,” a turn of phrase that neatly encapsulates the shift from computing as a virtual abstraction to computing as an engineered, physical commodity. With every hyperscaler seeking multi-gigawatt capacity pipelines, the North Dakota campus could become a case study in regional re-industrialization through AI infrastructure investment.
Why investors and analysts view Applied Digital as a bellwether for the next phase of AI-infrastructure growth
Applied Digital’s $5 billion lease represents not just corporate success but a potential blueprint for financing AI growth sustainably. Long-term, fixed-price leases offer capital-market visibility similar to power-purchase agreements in the renewable-energy sector, enabling developers to raise debt or equity against predictable cash flows. The company’s announcement may thus accelerate broader adoption of the AI-factory leasing model across North America.
Investors are watching whether Applied Digital can translate contracted backlog into realized EBITDA without dilution or cost overruns. The company’s ability to maintain margins amid construction inflation and power-market volatility will shape sentiment through 2026.
If Polaris Forge 2 delivers on schedule and within budget, Applied Digital could transition from a speculative small-cap to a benchmark AI-infrastructure franchise—a transformation that would validate both its strategic foresight and the Midwest’s viability as the next frontier for hyperscale AI deployment.
Discover more from Business-News-Today.com
Subscribe to get the latest posts sent to your email.
