TeraWulf Stock Jumps as Bitcoin Miner Secures Kentucky ‘Muskie’ Campus for 1+ GW of AI Capacity

Bitcoin miners are evolving into power-first data infrastructure developers, and TeraWulf just made that pivot louder. The company purchased a large development site in eastern Kentucky to build the Muskie Data Campus—positioning itself to deliver more than a gigawatt of AI and high‑performance computing capacity in a state hungry for new industrial investment.

Here’s the build TeraWulf mapped out. The campus—acquired from Industrial Equity Partners—sits inside the 1,000‑acre EastPark Industrial Park and includes roughly 285 acres of owned and controlled land, with optional adjacent parcels for future expansion. The land is already zoned for the intended use. Management targets the first 500 megawatts coming online in the second half of 2028, followed by another 500 megawatts by 2030. For scale, about one gigawatt can power roughly 750,000 homes.

Grid access is the linchpin. Kentucky Power is constructing a 345‑kilovolt substation connected to an existing 765 kV transmission network to serve the campus—exactly the kind of high‑voltage backbone AI data centers require. TeraWulf’s CEO Paul Prager framed the reality clearly in a statement: the binding constraint in this cycle isn’t chips; it’s energy, transmission, and reliable execution.

Public markets took notice. WULF shares climbed about 9% after the announcement to $24.78, hitting an intraday high of $25.92—its highest level in the past 12 months, per Yahoo Finance. The stock has more than doubled since the start of the year.

Strategically, this is TeraWulf’s second major Kentucky footprint, joining the 480‑megawatt Justified Data campus in Hancock County. Like peers IREN, MARA Holdings, and Hive Digital Technologies, TeraWulf is leaning into AI compute demand on the back of its existing power and real‑estate stack. That pivot is already visible in the numbers: in Q1, the company’s AI compute revenue surpassed its Bitcoin mining revenue, even as it posted a $427 million net loss.

The key to watch is power optionality. Owning and controlling land tied to a 765 kV backbone—and having a utility‑led 345 kV substation underway—confers a scarce asset: credible interconnection at scale. In today’s interconnection‑queue environment, sites that are pre‑zoned, adjacent to transmission, and supported by a willing utility can compress timelines and reduce variance in capex schedules. That matters for AI buyers who prioritize uptime assurances and for miners who need flexible load management to arbitrage power prices and market cycles.

There are trade‑offs. The schedule—H2 2028 for the first 500 MW and 2030 for the second tranche—pushes monetization into a future power market that could look very different. Capital intensity, long‑lead electrical gear, permitting cadence, and local workforce development all introduce execution risk. Energy mix remains a consideration, too; as Kentucky scales industrial load, how incremental generation is sourced will shape both operating costs and community outcomes. Miners have argued their flexible demand can stabilize grids; AI customers typically need steadier baseload, which may reduce curtailment opportunities but improve revenue predictability.

What I’m watching next: - Substation milestones and interconnection queue progress—slippage here ripples through the entire roadmap. - Long‑term power agreements or hedges that clarify cost structure and exposure to merchant volatility. - The split between third‑party AI tenants versus self‑operated compute, which will dictate margin and capital needs. - Financing terms against a backdrop of a stock that has rallied sharply but follows a quarter with a $427 million net loss.

If electricity is the currency of AI, miners with credible power footprints have a seat at the table. TeraWulf’s Muskie campus is a bet that control over electrons—and the transmission to move them—is the real moat.