Montana-Dakota Utilities Announces Electric Service Agreement with Applied Digital for Proposed AI Factory

Big promises, but little concrete progress or financial detail—watch, don’t chase yet.

What the company is saying

The company is positioning this announcement as a major step forward in building out critical infrastructure for the AI economy, specifically through a new electric service agreement (ESA) between Montana-Dakota Utilities (a subsidiary of NYSE:MDU) and Applied Digital (NASDAQ:APLD) for the Polaris Forge 3 project. Management wants investors to believe that this project will be transformative, citing the massive 430 megawatt power requirement and a 15-year lease with a 'high investment-grade hyperscaler' as evidence of strong demand and long-term stability. The language is assertive and future-focused, emphasizing anticipated benefits such as 200 new full-time jobs, meaningful property tax revenue, and regional economic growth. The announcement highlights the scale and ambition of the project, the prior successful collaboration at Polaris Forge 1, and the $38.4 million credited back to customers as proof of operational credibility. However, it buries or omits key financial details: there is no disclosure of the ESA’s dollar value, total project investment, or any near-term revenue or profit impact. The tone is upbeat and confident, with management projecting a sense of inevitability about the project’s success, but the communication style is more promotional than evidentiary. Notable individuals such as Wes Cummins (Chairman and CEO of Applied Digital) and Nicole Kivisto (President and CEO of MDU Resources) are named, lending institutional credibility, but their involvement is standard for a deal of this type and does not imply outside validation or third-party capital. This narrative fits into a broader investor relations strategy of associating the company with high-growth, future-facing sectors like AI and digital infrastructure, while leveraging regional economic impact as a secondary selling point. Compared to prior communications (where available), the messaging here is even more forward-looking and aspirational, with less emphasis on realised milestones and more on projected outcomes.

What the data suggests

The disclosed numbers are almost entirely project-specific and forward-looking, with little to no period-over-period financial data. The headline figure is the 430 megawatt power requirement for Polaris Forge 3, which signals a very large-scale data center but does not translate directly into revenue or profit without further detail. The only historical financial figure is the $38.4 million credited back to North Dakota customers over three years, but this is a utility rebate, not a measure of profitability or growth for either company. There is no disclosure of the ESA’s value, Applied Digital’s capital commitment, or any expected return on investment. The 15-year lease with a 'high investment-grade hyperscaler' is mentioned, but the tenant is unnamed and no lease rates or revenue projections are provided. The anticipated 200 full-time jobs and property tax impact are cited as economic benefits, but again, no supporting data or timeline is given. There is no information on capital expenditures, funding sources, or construction milestones. An independent analyst would conclude that while the project is ambitious, the lack of financial transparency and absence of binding, near-term milestones make it impossible to assess the true economic impact or risk profile. The gap between the company’s claims and the evidence is significant: the narrative is built on future potential, not current performance or secured contracts.

Analysis

The announcement is framed with a positive tone, highlighting a new electric service agreement and the anticipated benefits of a large-scale AI data center project. However, most of the key claims are forward-looking, including the projected start of operations in August 2027, expected job creation, and economic impact, with only a few realised facts (e.g., prior customer credits, existing service at another site). The benefits are long-dated, with initial operations not expected for over three years, and the scale of the project (430 MW) implies significant capital intensity, yet no immediate earnings impact or committed capital outlay is disclosed. The language inflates the signal by emphasizing anticipated outcomes and regional benefits without providing concrete, near-term financial or operational milestones. The data supports that an agreement has been entered into and that there is a history of collaboration, but lacks detail on binding commitments, regulatory approvals, or financial specifics for the new project. The gap between narrative and evidence is moderate: the announcement is more aspirational than milestone-driven.

Risk flags

• ●Execution risk is high: The project’s benefits are all contingent on successful completion of a large-scale, capital-intensive buildout, with initial operations not expected until August 2027. Delays or cost overruns are common in projects of this scale, and no construction start date or funding plan is disclosed.
• ●Regulatory risk is material: The electric service agreement and related filings require approval from the North Dakota Public Service Commission. There is no indication that these approvals are imminent or assured, and any delay or rejection would halt the project.
• ●Financial disclosure risk is significant: The announcement omits key financial metrics such as the value of the ESA, total project investment, expected returns, or funding sources. This lack of transparency makes it impossible for investors to assess the true risk/reward profile.
• ●Forward-looking bias: The majority of claims are aspirational and project-based, with little evidence of binding commitments or near-term milestones. This pattern increases the risk that the narrative is being used to generate hype rather than report progress.
• ●Customer concentration risk: The 15-year lease is with a single, unnamed 'high investment-grade hyperscaler.' If this tenant withdraws or delays, the entire project’s economics could be undermined.
• ●Capital intensity risk: The 430 megawatt requirement signals a massive capital outlay, but there is no detail on how this will be financed or what the impact on Applied Digital’s or MDU’s balance sheet will be. High capital intensity with distant payoff increases the risk of dilution or debt.
• ●Timeline risk: With initial operations not expected until August 2027, investors face a long wait before any potential payoff. The further out the timeline, the greater the risk that market conditions, technology, or customer needs will change.
• ●Omission of competitive or market context: The announcement does not address potential competition, market demand for AI data centers in the region, or alternative projects. This omission leaves investors blind to broader industry risks.

Bottom line

For investors, this announcement is more a statement of intent than a concrete, actionable milestone. The companies are signaling ambition and a desire to be seen as leaders in the AI infrastructure space, but the lack of financial detail, binding commitments, or near-term milestones means there is little to anchor the narrative in reality. The involvement of senior executives like Wes Cummins and Nicole Kivisto is expected for a deal of this size, but does not provide additional validation or guarantee of execution. To change this assessment, the companies would need to disclose signed, binding agreements for project funding, construction, or offtake, as well as secure regulatory approvals and provide a clear, detailed construction timeline. Investors should watch for regulatory filings, construction start dates, and any updates on project financing or tenant commitments in the next reporting period. At this stage, the announcement is a weak positive signal—worth monitoring, but not strong enough to justify new investment or a change in position. The most important takeaway is that while the project could be transformative if executed as described, the path to realization is long, uncertain, and currently unsupported by hard financial evidence. Treat this as a watchlist item, not a buy signal.

Announcement summary

(NYSE: MDU) MDU Resources Group, Inc.'s subsidiary, Montana-Dakota Utilities Co., has entered into an electric service agreement (ESA) with Applied Digital Corporation (NASDAQ: APLD) to provide power to Polaris Forge 3, an AI Factory near Center, North Dakota. At full capacity, the campus would require 430 megawatts of electricity. Applied Digital anticipates initial operations to commence in August 2027. Applied Digital has previously announced a 15-year lease with a U.S. based high investment-grade hyperscaler for this site. The campus is expected to create approximately 200 full-time jobs and generate meaningful property tax revenue. Montana-Dakota Utilities currently serves Applied Digital at Polaris Forge 1, where $38.4 million has been credited back to North Dakota customers over the past three years. Approval of the ESA and other regulatory filings by the North Dakota Public Service Commission is required for the company to provide power under the agreement.

Disagree with this article?