Applied Digital's $16 Billion Contract Mirage: Or, How to Read Between the Lines
Alright, let's talk about Applied Digital (APLD). The headline grabbing number? A cool $16 billion in contracted revenue across its Polaris Forge campuses. Sounds impressive, right? Like they’ve struck AI gold. But before you jump on the bandwagon, let's dig into what that number really means.
The company is touting the completion of Phase II at its Polaris Forge 1 campus in Ellendale, North Dakota, bringing the energized capacity to 100 MW. This is part of a 400 MW deployment for CoreWeave under long-term lease agreements. They're building AI factories, essentially, and claiming they're doing it faster and more reliably than anyone else. Applied Digital Completes 100 MW at 400 MW Polaris Campus
Then there's the $2.35 billion offering of senior secured notes due in 2030. The stated purpose? To fund construction of two more data centers (ELN-02 and ELN-03) at the same Polaris Forge 1 campus, repay existing debt, and cover transaction expenses.
So, we have massive revenue projections and massive debt being taken on to supposedly get there. That's the setup.
Here’s where things get interesting. That $16 billion figure? It's "contracted revenue." Not actual revenue. It's what they expect to make over the entire life of these long-term lease agreements. It's like saying you're worth a million dollars because you have a 30-year mortgage and plan to work until you're 70.
Applied Digital is essentially betting big on the continued, exponential growth of AI compute demand. And maybe they’re right. But "betting big" is the key phrase here.
The company also highlights that they saw 243 institutional investors add shares of $APLD stock to their portfolio, while 155 decreased their positions in their most recent quarter. On the surface, that's positive, but the size of those moves matters. Were the additions larger than the reductions? The data doesn't say. (And that's a pretty crucial omission, isn't it?)
And here's a detail that made me raise an eyebrow: insiders have been selling stock. A lot of stock. Over the past six months, insiders have sold shares 18 times, with zero purchases. CEO Wes Cummins sold 400,000 shares for an estimated $6.1 million. CFO Mohammad Saidal Lavanway Mohmand sold 212,503 shares for nearly $6 million. (I've looked at hundreds of these filings, and that level of consistent selling always makes me wonder.) It's not necessarily a red flag (founders often diversify their holdings), but it is a data point that needs to be considered alongside everything else.
The analyst ratings are overwhelmingly positive – a bunch of "buy" or "outperform" ratings with a median price target of $37. But analysts, let's be honest, often lag behind the curve. And their price targets are based on models that, like Applied Digital's revenue projections, rely on a whole lot of assumptions about the future.
Applied Digital is building these campuses, taking on billions in debt, and projecting massive revenues based on long-term contracts in a rapidly evolving market. The phrase "high-risk, high-reward" comes to mind, but even that sounds a bit too generous.
The proposed offering of $2.35 billion in senior secured notes is the linchpin here. They're using that money to build more data centers to fulfill those long-term contracts. If the demand for AI compute falters, or if they can't deliver on time and within budget, those notes could become a very heavy weight around their neck. Applied Digital Announces Proposed Offering of $2.35 Billion of Senior Secured Notes
This reminds me of the dot-com boom. Companies were building fiber optic networks based on projections of internet bandwidth that seemed reasonable at the time. But technology changed, demand shifted, and a lot of those companies went bust, leaving behind miles of unused fiber. Applied Digital is betting that AI is different. That the demand is real, sustainable, and predictable.
The company is essentially selling shovels in a gold rush. And like any gold rush, some people will get rich, and a lot of people will end up holding empty pans.
The data paints a picture of a company making bold moves in a high-growth, but also high-risk, sector. That $16 billion in contracted revenue is more of a potential than a sure thing. And the insider selling, while not definitive, is a detail that can't be ignored. It doesn't mean $APLD stock is necessarily a bad investment, but it absolutely means you need to do your homework and understand the risks before jumping in. The truth isn't in the press release; it's in the numbers, and in what the numbers don't say.
