πŸ“° Key Highlights

Crypto infrastructure companies are continuing to expand their Texas land footprint, driven by growing power demand from AI and high-performance computing (HPC). Galaxy Digital announced on Tuesday that it has purchased a 500-acre parcel in McGregor, Texas, where it will build its second AI and HPC data center campus in the state. The initial phase is planned for 74 megawatts of power capacity, with phased expansions to follow. On the same day, Bitcoin miner MARA Holdings announced an acquisition agreement for a 1,200-acre site in Matagorda County with existing power access, supporting up to 2 gigawatts (2 GW) of capacity. The company said it will develop the site for AI and HPC workloads while also using it as a Bitcoin mining base. Also on that day, Meta disclosed plans to build a $14 billion AI data center campus in El Paso, Texas, further fueling the wave of large-scale infrastructure investment in the state. Texas has become a popular location for data center and crypto infrastructure projects primarily because of its large power market, relatively low electricity costs, and ample land available for large-scale developments. Additionally, Texas’s independent grid operator ERCOT has attracted many developers needing substantial power capacity.

πŸ’¬ JudyAI Lab Perspective

Crypto miners and cloud giants in Texas are doubling down on AI and HPC data center builds β€” land and power have become the new battleground for the next wave of infrastructure.

On the same day, Galaxy Digital purchased 500 acres in McGregor to build its second Texas AI/HPC campus, MARA Holdings secured 1,200 acres of powered land in Matagorda County for a dual-use AI and mining site, and Meta announced a $14 billion AI data center project in El Paso. This reflects a clear trend: compute demand is growing faster than the existing power grid can expand. Whoever locks in power capacity and land first holds the entry ticket to the next phase of AI infrastructure. Texas, with its low electricity costs, large power market, ample land, and the ERCOT independent grid, has become a key node in this race. It also shows that AI competition is extending from model capabilities to positioning battles over physical resources like power and land.

For AI builders, it’s worth noting that not only model technology matters β€” supply bottlenecks in infrastructure will also shape the industry’s pace and cost structure.

πŸ“… Source Info

πŸ”— Further Reading