Galaxy Digital's 74-MW Texas AI Campus Could Still Be a Bargain-if the Build Keeps Passing the Smell Test

Galaxy Digital's recent slide shifts the debate to execution

Galaxy Digital is now a before-or-after stock after trading down 21.8% over the past year. That weakness matters because the market is no longer paying for a simple AI headline. It is pricing a bet on whether the approximately 500 acres in McGregor becomes a real, powered, leased campus-or just another big land story with a long leash.

Bulls see execution; bears see funding risk

Bulls can point to a tangible pipeline. Galaxy lists 1.6+ GW Approved Data Center Power Capacity in its data center business, and McGregor is past the announcement phase with construction underway on the first phase. That makes it easier to evaluate than a purely theoretical project.

Bears do not necessarily dispute AI demand. Their concern is whether Galaxy can fund and operate a capital-heavy build without stretching the balance sheet. The real debate is whether power, financing, and customers will arrive in a clean enough sequence to support the thesis.

McGregor looks more credible than a typical land announcement

After the recent 18.8% drop over the past 30 days, the question is less about storytelling and more about whether this looks like a real industrial build. On that score, the setup is more credible than a standard land announcement. The clearest proof point is construction underway on the first phase. Once work is active on site, investors can track whether progress is sustained.

Power interconnection matters more than acreage

The main bottleneck in data centers is not land; it is getting power built on time. That is why progress on interconnection facilities matters so much. It also matters that the initial phase is expected to reach 74 MW, which gives investors a concrete checkpoint rather than a vague long-term promise.

Galaxy's private-funding commitments matter

Galaxy has also said it will privately fund an on-site electrical substation and cover additional water infrastructure. That does not prove end demand, but it does show the company is putting capital behind the grid and local constraints instead of waiting for utilities to do all the heavy lifting.

Galaxy already has a Texas buildout reference point

Galaxy also has a larger Texas benchmark to reference: its 1.6 GW Helios development. That does not guarantee McGregor will fill on schedule, but it does show the company is not starting from zero. For now, the clearest test is whether construction, grid work, and timeline updates keep staying concrete.

The bear case is financing discipline, not AI demand

The bear case is not about whether Texas needs more compute. It is about whether Galaxy can fund the build without stretching itself too thin.

Real demand can exist even if the stock still fails

Bears separate two questions: whether demand for AI infrastructure exists, and whether Galaxy has the financial strength to turn that demand into a finished, leased asset. That is the real controversy here. Even if demand is strong, Galaxy is still betting on a multi-hundred-megawatt campus through 2030 built off an initial development phase expected to reach 74 MW. Large projects of this kind can consume cash long before revenue catches up.

Cost pressure can arrive before customer de-risking

McGregor is not just a paper plan. It comes with construction underway on the first phase and commitments to privately fund an on-site electrical substation, provide financial assurances for utility upgrades, and pay for additional water infrastructure. That kind of structure can still create cost pressure before customers fully de-risk the project.

Galaxy's hybrid business adds another layer of risk

Galaxy also describes itself as a digital assets and AI infrastructure company with $8B Assets on Platform and a $1.4B Average Loan Book Size. That cuts both ways: the company understands capital markets, but it also raises the risk that AI buildout exposure gets mixed into a broader balance-sheet story. Bulls can argue McGregor fits a multi-campus strategy; bears can argue a busier construction site is not the same thing as a stronger credit profile or a firmer customer backlog.

What would make GLXY more buyable?

The question now is not whether interest exists. It is whether execution keeps improving.

A practical watchlist

• Stay with the clearest signal first: is construction underway on the first phase still progressing, or has activity stalled?
• Watch power, not promises: track progress on interconnection facilities and whether Galaxy honors its privately funded substation commitment.
• The next big step is customer proof: watch for updates on power contracts, and customer commitments.

What would weaken the setup

• If site activity fades, funding commitments slip, or timeline updates grow vague again, the bear case gets stronger.
• The bigger risk is financing discipline. Galaxy is already pursuing a multi-campus strategy, and that gets more stressful if one project starts pulling resources from the rest.

This looks more like a watch-then-buy setup than a blind-faith story. For now, the cleaner approach is to wait for physical progress, grid progress, and customer progress to line up. If they do, the market may start to treat McGregor as a real AI build rather than an ambitious land plan.