Applied Digital's New Credit Line Says AI Has Reached The Bank Desk

TL;DR: Applied Digital's new credit line matters less as another AI-infrastructure headline than as a financing signal. The June 8 facility gives the company up to $550 million of borrowing capacity, but the bigger point is that AI data-center developers are now being judged like project-finance operators, not just fast-growing capex stories.

The scene that matters is not a GPU rack. It is a lender desk.

A banker is looking at a partially leased campus, a construction calendar, a tenant-credit file, and a working-capital need that arrives before the full rent stream does. That is why Applied Digital's new revolver looks important. It is a reminder that the AI buildout is no longer just about who can announce megawatts. It is about who can carry the ugly middle period between lease signing and stabilized cash flow.

#The market keeps talking about AI demand. The financing clock is the real story.

Applied Digital said Monday that it closed a revolving credit facility arranged by Goldman Sachs with up to $350 million of committed capacity and another $200 million accordion. The money is meant for pre- and post-lease development and working capital. That wording matters.

This is not glamour financing. It is bridge financing for the awkward phase where a data-center developer has signed attractive leases, but still needs to fund site work, fit-out sequencing, vendor payments, and the normal mess of getting a project from promising to bankable.

The market still prices many AI-infrastructure names as if the hard part is winning demand. Demand is not the scarce resource anymore. Balance-sheet stamina is.

#The quiet clue is the collateral, not the headline size.

Applied Digital said the facility is secured by certain non-data-center project assets and matures in May 2029 at SOFR plus 225 basis points or base rate plus 125 basis points. That is a very different message from the loose, equity-funded AI boom story many investors still carry around.

It says lenders are willing to support the company, but they still want clean structures, real security, and defined sources of repayment. In other words, AI infrastructure is starting to look more like power, pipelines, and transport assets than like a normal software growth trade.

That shift has two consequences:

Equity investors will have to care more about financing mix, lease quality, and timing gaps, not just contracted revenue slides.
Developers with decent access to credit may widen the gap over peers that have impressive customer logos but weak funding flexibility.

#Why the CoreWeave clause is the most revealing part

The most interesting line in the release was not the revolver itself. It was the side note that Applied Digital signed a June 5 memorandum with CoreWeave to assign the Building 3 lease at Polaris Forge 1 to a CoreWeave subsidiary if that subsidiary reaches investment-grade status.

That sounds technical. It is actually the whole game.

If tenant credit improves, the value of the lease improves. If the lease improves, project debt gets cheaper. If project debt gets cheaper, the developer keeps more equity upside. This is how an AI data-center story turns into a financing-arbitrage story.

Applied Digital's own materials have been pointing in this direction for months. Its April investor deck says Polaris Forge 1 has three CoreWeave-related leases covering 400 MW and about $11 billion of anticipated contract value over 15 years, while the April earnings release said management expected credit enhancements tied to CoreWeave's refinancing to help lower overall debt costs over time and turn these campuses into stabilized, cash-flowing assets.

That is the hidden mechanism most casual readers miss. The upside is not just renting space to AI tenants. The upside is converting those leases into cheaper capital.

#What investors are probably still mispricing

I think the market is still mixing together two very different AI-infrastructure businesses.

One group is selling the picks and shovels: chips, networking gear, power equipment. The other group is trying to turn long-duration AI leases into an asset-backed financing machine. Applied Digital is pushing hard into the second category.

That means the main questions are changing:

Can it keep lowering its cost of debt as projects mature?
Can tenant-credit quality keep improving fast enough to re-rate the financing stack?
Can it fund the build cycle without giving away too much equity in the process?

Those are not side questions. They are the business model.

If this logic holds, the winners in the next leg of the AI buildout will not just be the companies with the biggest campuses. They will be the ones that convince banks and bond buyers that those campuses deserve infrastructure-style financing.

That is a much stricter test than getting investors excited about another lease announcement. The AI trade is growing up, and adulthood usually arrives in the credit documents.

#FAQ

Why does a revolving credit facility matter if the company already has long-term leases?

Because signed leases do not eliminate the timing gap between development spending and fully stabilized rent collection. A revolver helps fund that messy middle.

Why is the CoreWeave investment-grade condition such a big deal?

Because better tenant credit can lower financing costs for the project. In this business, the quality of the tenant can change the cost of capital almost as much as the quality of the building.