Applied Digital’s AI Data Center Backlog Could Multiply Its Stock, but Execution Risks Remain

Applied Digital reached Google News with a bold prediction: its AI infrastructure contracts will turn the stock into a multibagger within three years. That claim followed a fiscal year when revenue rose 167%, while fourth-quarter revenue increased roughly fivefold from the prior year. The numbers demand attention, but they do not settle the investment case.

The prediction appeared in an August 1 article distributed by The Motley Fool and AOL. Its argument centers on Applied Digital’s signed leases, expanding construction pipeline, and expected conversion of contracted capacity into recurring revenue. The company now reports 1.4 gigawatts of contracted critical IT load across five planned AI campuses.

Applied Digital is not competing directly with Nvidia in chip design or with Microsoft in cloud software. Its real contest is between contracted demand and execution risk. The company must finance, build, energize, and deliver specialized data centers before its backlog becomes operating cash flow.

That distinction matters because a multibagger describes a stock that rises several times above an investor’s original purchase value. It is an outcome, not a business metric. Neither lease announcements nor projected revenue can guarantee it.

What the Google News Prediction Actually Claims

The bullish argument treats Applied Digital’s lease backlog as evidence that its current revenue base understates the company’s future scale.

The original multibagger prediction identifies Applied Digital as a focused way to gain exposure to AI infrastructure spending. Applied Digital designs, develops, and operates high-density data centers for artificial intelligence and high-performance computing customers.

The thesis begins with the size of the broader build-out. Goldman Sachs estimates cited in the article suggest that AI data centers, computing hardware, and electricity spending might reach $7.6 trillion between 2026 and 2031. That forecast is a scenario, not committed spending.

Applied Digital’s company-specific numbers provide the stronger foundation. Management says signed agreements cover 1.4 gigawatts of critical IT load and approximately 2.15 gigawatts of grid-connected utility power. Critical IT load represents the electricity available to computing equipment, excluding supporting systems such as cooling.

The company also reports about $36 billion of contracted base-term lease revenue. That figure reaches $86 billion if customers exercise every available renewal option. Investors should separate those categories because optional renewals are not equivalent to base-term commitments.

The portfolio spans five AI-focused campuses. Applied Digital says approximately 70% of its contracted revenue comes from investment-grade U.S. hyperscalers. A hyperscaler is a large cloud or technology company that operates computing infrastructure at enormous scale.

Its newest disclosed agreement covers 210 megawatts at the Delta Forge 2 campus. The 15-year take-or-pay lease represents approximately $5.2 billion of base-term contracted revenue. Take-or-pay terms generally require the customer to pay for reserved capacity once contractual conditions are satisfied.

Applied Digital’s lease announcement says the same agreement reaches approximately $12.7 billion if every renewal option is used. The site is designed for large AI training and inference workloads, according to the company.

These commitments look enormous beside Applied Digital’s current operations. Fiscal 2026 revenue was approximately $611 million, according to results cited by the original analysis. Fourth-quarter revenue reached approximately $259 million as completed capacity began producing lease revenue.

The prediction assumes that additional facilities will repeat that conversion process. Construction becomes completed capacity, completed capacity begins a lease, and the lease produces recurring revenue over many years.

That mechanism is plausible. However, it contains more dependencies than a simple comparison between backlog and current sales suggests.

Applied Digital must secure construction capital, maintain utility schedules, satisfy tenant specifications, and reach contractual readiness dates. It must accomplish those steps across several campuses without allowing costs or delays to overwhelm expected returns.

The Google News headline compresses this chain into a stock-price conclusion. The underlying story is more useful when read as an execution test. Applied Digital has already established demand, but it has not yet converted most contracted demand into delivered infrastructure.

Why Applied Digital Is Winning AI Data Center Contracts

Applied Digital’s appeal comes from its ability to pair large power positions with facilities designed around dense AI computing systems.

AI data centers differ from conventional enterprise facilities because modern accelerator clusters concentrate substantial power and heat inside each rack. They also need fast networking between thousands of processors. General-purpose buildings often require significant redesign before they can support those workloads.

Applied Digital markets campuses built specifically for AI training and inference. Training creates or updates a model using large datasets. Inference uses a trained model to answer requests or generate outputs.

The company says its designs include high-density power infrastructure and waterless cooling. Waterless cooling reduces reliance on evaporative systems, although the complete environmental effect still depends on energy sources and local conditions.

Power availability has become one of the industry’s central constraints. A developer can acquire land and order equipment, but a large campus cannot operate without generation, transmission, substations, and utility approval.

Applied Digital has pursued sites where it believes those resources can support several hundred megawatts. Its North Dakota campus provided the earliest evidence that the model could attract a major AI tenant.

In June 2025, Applied Digital announced two leases with CoreWeave covering 250 megawatts. The agreements had initial terms of roughly 15 years. One building was planned for 100 megawatts, while another was planned for 150 megawatts.

The related CoreWeave filing confirms the long initial terms and three five-year extension options. It also shows that these are detailed property agreements, not informal expressions of interest.

CoreWeave operates a specialized cloud platform built around graphics processing units, or GPUs. GPUs perform many calculations in parallel, which makes them useful for training and serving AI models.

The relationship gave Applied Digital a significant anchor customer. It also revealed a concentration risk because a smaller developer can become dependent on a limited number of tenants and their financing capacity.

Applied Digital later expanded beyond those initial CoreWeave leases. Its 2026 announcements described additional agreements with unnamed investment-grade hyperscalers across new campuses.

That shift supports the bullish case in two ways. First, it increases contracted capacity. Second, it reduces the percentage of the backlog tied exclusively to CoreWeave.

The company’s model also offers customers an alternative to owning every data center. A technology company can reserve purpose-built capacity through a long lease while leaving land development, construction, and operations to Applied Digital.

This division of labor has clear appeal during a supply shortage. Hyperscalers want capacity quickly, yet internal projects can face the same power and construction constraints as outside developers.

Applied Digital’s potential advantage is therefore not a unique AI model or semiconductor. It is a portfolio of developable sites, utility relationships, engineering capabilities, and tenant contracts.

That advantage remains highly physical. Software can reach millions of users without constructing a new building for each customer. Applied Digital must invest at each site before revenue begins.

The company’s history also deserves attention. It previously developed infrastructure for blockchain computing before shifting its emphasis toward high-performance computing and AI.

That experience supplied knowledge about power-intensive facilities. It does not automatically prove the company can deliver several large AI campuses at the same time. AI tenants demand different equipment, cooling arrangements, redundancy, and construction standards.

Still, signed leases indicate that sophisticated customers have reviewed the plans. Those customers are not guaranteeing Applied Digital’s stock returns, but their commitments validate demand for the proposed capacity.

The most credible bullish conclusion is therefore narrower than the headline prediction. Applied Digital has assembled a sizable contracted pipeline in a market where access to power and timely delivery have become valuable.

The Real Contest Is Backlog Versus Execution

Applied Digital’s multibagger case succeeds only if signed revenue becomes completed, financed, and profitable capacity.

Backlog can provide visibility, but it is not the same as reported revenue. Accounting recognition usually begins after a facility reaches the required condition and the customer takes control of its leased capacity.

Before that point, Applied Digital must spend money on land, buildings, electrical systems, cooling equipment, and network infrastructure. It may also need tenant deposits, project debt, partner capital, or corporate financing.

This creates a timing gap. Large contractual values arrive in headlines immediately, while much of the related revenue arrives over 15 years. Construction costs and financing needs arrive earlier.

The original prediction uses a scenario where annual revenue reaches approximately $2.7 billion in fiscal 2029. It then applies a sales multiple near the broader U.S. technology sector average.

That calculation produces a possible market value almost three times the company’s value when the article appeared. It is a transparent scenario, but every major input can change.

Revenue might exceed the estimate if Applied Digital delivers early and signs more leases. It might fall short if construction milestones slip, energization takes longer, or tenants delay acceptance.

The valuation multiple creates another uncertainty. Applied Digital owns and develops capital-intensive infrastructure. Investors may not value it like a high-margin software company, even if its growth rate resembles one temporarily.

A sales multiple also says little about interest expense, preferred distributions, depreciation, maintenance capital, or shareholder dilution. Those factors determine how much economic value reaches common shareholders.

Applied Digital’s financing arrangements show why this matters. An investor presentation filed with the SEC describes preferred capital carrying a 12.75% distribution during its early years. The structure also includes attached equity and later rate increases.

The financing terms demonstrate access to institutional capital. They also show that infrastructure funding is not free.

Applied Digital can create value even with expensive capital if tenant revenue substantially exceeds project costs. However, headline lease values alone cannot establish that spread.

Construction performance presents a separate risk. Management says it has delivered capacity on time and within budget. Investors should treat that as a company claim and compare it with each future readiness date.

Large projects encounter weather, labor shortages, equipment lead times, permitting changes, and utility delays. A problem at one site might remain contained. Similar delays across multiple campuses would undermine the scale thesis.

Grid access deserves particular scrutiny. The company describes approximately 2.15 gigawatts of grid-connected utility power across its contracted portfolio. That figure does not necessarily mean every megawatt is ready for immediate tenant use.

Substations, transmission upgrades, generation capacity, and commissioning schedules can determine when a campus becomes operational. Investors need precise service dates rather than broad power totals.

Customer quality reduces some risk, but it does not eliminate dependency. Long leases can protect revenue once contractual obligations begin. They cannot prevent every amendment, dispute, redesign, or delay before service starts.

Applied Digital and CoreWeave amended parts of their lease structure in March 2026. The lease amendment included suspended terms for two data halls and a new agreement with another CoreWeave entity.

That change did not erase the broader commercial relationship. It illustrates why investors must read the filings behind aggregate contract figures. Individual leases can evolve as financing and deployment plans change.

The bear case does not require AI demand to collapse. Applied Digital can face disappointing shareholder returns even while the industry grows.

Costs might rise faster than rents. New shares might dilute existing owners. Interest expenses might absorb operating income. Customers might shift deployment schedules as chips become more efficient.

The bullish argument must therefore clear a higher bar than proving AI data centers are needed. It must show that Applied Digital can earn attractive returns after financing and construction costs.

This is the main reversal hidden beneath the Google News headline. A historic build-out creates demand, but it also creates an enormous capital requirement for the companies supplying capacity.

CoreWeave, Hyperscalers, and the Pressure on Capacity Providers

Applied Digital occupies a useful middle position, but larger customers and better-capitalized rivals retain considerable bargaining power.

The company sits between utilities, capital providers, equipment suppliers, and AI infrastructure customers. Each participant can affect a project’s economics or completion date.

CoreWeave helped establish Applied Digital’s AI data center strategy. It also brings its own financial and operational pressures because specialized cloud companies must keep buying accelerators while expanding facilities.

A long-term lease transfers some property-development work from CoreWeave to Applied Digital. It does not disconnect either company from end-user demand for AI computing.

If enterprises continue expanding AI workloads, CoreWeave needs more capacity. If deployment growth slows, every infrastructure provider will face closer scrutiny of commitments and utilization.

Unnamed investment-grade hyperscalers offer a different risk profile. Their stronger balance sheets can make a lease easier to finance. Their size can also give them significant negotiating leverage.

Applied Digital competes with several routes to capacity. Hyperscalers can build their own campuses, lease from established data center operators, or use emerging developers with available power.

Digital Realty and Equinix represent mature data center operators with global portfolios. Their scale, customer diversity, and financing histories differ greatly from Applied Digital’s focused development strategy.

Core Scientific and other former cryptocurrency infrastructure operators have also pursued AI hosting opportunities. These companies understand high-density power operations, but conversions can require extensive work.

Developers such as Applied Digital can move faster when they control attractive power sites. Established operators may offer lower perceived execution risk and wider geographic choice.

Hyperscalers can also change facility requirements as hardware evolves. A campus designed for one generation of accelerators may need altered cooling or network layouts for later systems.

Applied Digital’s long leases help address this risk through committed terms. Yet contract protection does not guarantee that every building will remain equally competitive throughout its useful life.

The industry’s capacity race also creates supply-chain competition. Data center developers need transformers, switchgear, generators, cooling systems, and skilled construction labor.

When many companies order the same equipment, lead times and prices can rise. Larger buyers may secure priority because they place recurring orders across multiple markets.

Applied Digital’s franchise model attempts to standardize development across campuses. Management says a central design, construction, and operations team can reproduce its approach in new locations.

Standardization should reduce repeated engineering work. Regional differences still matter because each site has distinct utilities, permits, weather conditions, labor markets, and community concerns.

Local opposition can also delay projects. Data centers may create tax revenue and construction employment, but communities increasingly question their electricity and water demands.

Applied Digital emphasizes waterless cooling in its latest campus announcement. That design can answer one concern, while power sourcing and transmission impacts remain relevant.

Developers also face a technological question. AI hardware has historically improved its output per unit of energy, but total demand can still rise because cheaper computation encourages more use.

This is known as a rebound effect. Efficiency reduces the resources needed for one task, while growing adoption increases the total number of tasks.

If total demand keeps rising, Applied Digital’s power pipeline gains strategic value. If efficiency outpaces adoption, some planned capacity might become less urgent.

The market currently rewards evidence that capacity has both power and a tenant. Applied Digital has accumulated both on paper. The next stage requires physical delivery at several sites.

That stage will reveal whether its smaller size enables faster decisions or creates operational strain. It will also determine whether the company can negotiate future leases without weakening project returns.

What the Multibagger Math Leaves Out

The stock-price prediction depends on valuation assumptions that are less certain than the company’s signed lease terms.

The original analysis presents a straightforward path. It starts with projected fiscal 2029 revenue near $2.7 billion, applies an 8.7 price-to-sales multiple, and arrives at a market value near $23.5 billion.

That result was almost three times Applied Digital’s reported market value when the article ran. The calculation explains the multibagger label, but it should not be mistaken for an independent forecast.

Analyst revenue estimates can change as projects move through construction. Sales multiples can also compress even when revenue grows, especially if investors demand stronger cash flow.

The choice of comparison group matters. A technology-sector average combines businesses with very different margins, capital requirements, and balance sheets.

Applied Digital’s recurring leases may justify a premium over ordinary construction companies. Its funding requirements may justify a discount to software businesses.

Investors should focus on four layers of economics. The first is total contracted rent. The second is the cost of building each campus.

The third is the financing cost attached to that construction. The fourth is the ownership share remaining for common shareholders after project partners and new equity are considered.

A lease can be economically attractive at the property level while producing modest returns for existing shareholders. This happens when financing partners claim a large portion of the project’s value.

Conversely, a developer can create substantial equity value when it secures land and power early, signs a strong tenant, and finances construction on favorable terms.

Applied Digital’s reported backlog shows that it has completed the tenant step across several campuses. Investors still need enough information to evaluate the other steps consistently.

Revenue growth alone will not answer that question. Funds from operations, adjusted earnings, and operating cash flow can each offer useful information, but none should be viewed alone.

Capital expenditures must remain part of the analysis. A company spending heavily to create long-lived assets will often report cash outflows before those assets generate rent.

Debt maturity schedules also matter. Refinancing during a weak credit market can reduce equity returns, even when the underlying buildings remain occupied.

Preferred securities can avoid immediate common-stock issuance, but they introduce distributions and claims ahead of common shareholders. Convertible debt can later become equity and increase the share count.

These are normal tools for infrastructure development. They become dangerous when a company builds too much speculative capacity or repeatedly finances projects on unfavorable terms.

Applied Digital has reduced speculative risk by signing long leases. Yet some campuses still require years of execution before their full revenue contribution appears.

The prediction also uses the phrase “easily become a multibagger.” That wording understates the difficulty of completing multiple large projects while preserving shareholder economics.

A more defensible statement is that Applied Digital has a credible path to much greater revenue. Whether the stock multiplies depends on margins, financing, dilution, and the valuation investors assign later.

The Google News audience should also distinguish reporting from recommendation. A widely distributed opinion article reflects one analyst’s scenario and assumptions.

It does not establish a consensus price target. It does not guarantee that the market will value future sales at the selected multiple.

The headline’s certainty makes the story clickable. The underlying facts support a conditional thesis, not an inevitable outcome.

That does not make the opportunity unimportant. It makes the relevant question sharper: can Applied Digital convert unusually large contractual demand into attractive per-share cash flows?

Three Signals That Will Test the Applied Digital Thesis

The next decisive evidence will come from delivery dates, project-level financing, and revenue conversion rather than additional headline forecasts.

The first signal is campus commissioning. Investors should track when each contracted building reaches its ready-for-service date and begins recognizing lease revenue.

A ready-for-service date marks the point when a facility satisfies agreed conditions for the tenant. Delays would push revenue into later periods and might increase construction or financing costs.

On-time delivery would strengthen management’s claim that its development process can scale. Repeated delays would weaken the multibagger thesis even if the underlying leases remain intact.

The most useful disclosures will identify individual facilities, contracted megawatts, scheduled service dates, and actual acceptance dates. Aggregate gigawatt figures cannot replace that project-level record.

The second signal is the financing structure for each new campus. Investors should examine interest rates, preferred claims, equity contributions, guarantees, and ownership arrangements.

A large funding announcement can sound positive while carrying expensive terms. The critical issue is how much expected project value remains available to Applied Digital’s common shareholders.

Lower financing costs would strengthen the investment case because long leases can support predictable project cash flows. Rising costs or repeated equity issuance would dilute the upside.

Investors should also compare project funding with construction commitments. Securing a tenant without enough capital to deliver the building would leave the central execution problem unresolved.

The third signal is the conversion of backlog into recurring revenue and cash generation. Applied Digital’s fourth-quarter acceleration shows that delivered infrastructure can materially change reported sales.

The important test is whether that process repeats across campuses. Revenue should rise as each facility begins its lease, while operating cash flow should improve after initial construction periods.

Margins deserve equal attention. Rapid revenue growth with weak project returns would challenge the assumption that scale produces shareholder value.

Customer concentration should appear beside those metrics. Additional investment-grade tenants would reduce dependence on any single cloud provider and improve the resilience of contracted revenue.

Renewal-option values should remain separate from base terms. Investors should not count the entire $86 billion as committed revenue because customers control those extensions.

Broader hyperscaler spending remains relevant, but it is not the best near-term test. Applied Digital already reports enough demand to support a much larger business.

Its constraint has moved from finding an AI narrative to delivering contracted infrastructure. That is why another industry forecast carries less information than an energized data hall.

Readers who encountered the prediction through Google News should treat it as the opening claim, not the conclusion. The article correctly identifies a company with extraordinary contracted growth potential.

However, the strongest evidence will arrive through filings, construction updates, and financial results. Those disclosures will reveal whether the backlog supports profitable scale or masks costly complexity.

Applied Digital has secured a place in the AI infrastructure race. It has long leases, major customers, and a development pipeline measured in gigawatts.

It also carries the obligations that accompany those assets. Each campus requires capital, equipment, grid coordination, and precise delivery.

The stock becomes a multibagger only if business growth reaches shareholders on acceptable terms. That outcome remains possible, but it is neither easy nor assured.

Over the next several quarters, ask three questions. Are contracted facilities entering service on schedule? Is Applied Digital financing them without surrendering excessive value? Is rising revenue producing stronger per-share economics?

Those answers will matter more than any prediction circulating through Google News.