Microsoft Acquires Texas Data Center Campus Abandoned by Oracle and OpenAI
In a strategic move to bolster its cloud infrastructure, Microsoft has acquired a sprawling data center campus in Abilene, Texas, that was previously slated for development by Oracle and OpenAI. The site, which spans over 877 acres and includes plans for up to 20 data center buildings, represents a significant expansion opportunity for Microsoft’s Azure platform amid surging demand for AI workloads.
The campus, located at 200 Global Drive in Abilene, was initially announced in May 2024 as the future home of a massive AI data center project jointly pursued by Oracle and OpenAI. Dubbed part of the broader Stargate initiative, the facility was envisioned to support OpenAI’s ambitious supercomputing needs, powered by a colossal 1.2 gigawatts of capacity. This power draw, equivalent to the output of a large nuclear power plant, underscored the scale of the proposed infrastructure, designed to fuel next-generation AI training and inference.
However, Oracle and OpenAI abruptly stepped away from the project earlier this year. Sources indicate that the partnership faced challenges, including regulatory hurdles, escalating construction costs, and shifts in strategic priorities. OpenAI, in particular, has been navigating complex funding arrangements with Microsoft, its primary backer, while exploring alternative sites for its AI ambitions. Oracle, meanwhile, continues to expand its own cloud regions but appears to have deprioritized this Texas location.
Microsoft’s acquisition, confirmed through local permitting records and property filings with the Taylor County Appraisal District, positions the company to repurpose the site for Azure hyperscale operations. The campus already benefits from approved zoning for data center use and proximity to high-voltage transmission lines, critical for high-density computing. Abilene’s selection aligns with Microsoft’s broader strategy of targeting West Texas for its abundant land, renewable energy potential, and lower electricity costs compared to coastal regions.
From a technical standpoint, the facility’s design specifications are tailor-made for AI-era demands. Each planned building could house hundreds of thousands of GPU racks, with liquid cooling systems essential for managing the thermal loads of Nvidia H100 or Blackwell GPUs. The 1.2GW power allocation would enable unprecedented compute density, supporting models with trillions of parameters. Microsoft’s expertise in Azure’s AI infrastructure, including integrations with its Maia chips and partnerships with Nvidia, equips it to operationalize the site swiftly.
This deal highlights intensifying competition in the hyperscale data center market. Providers like Microsoft, Amazon Web Services, and Google Cloud are racing to secure megasites capable of gigawatt-scale power, driven by the exponential growth in AI training requirements. A single GPT-4 level model training run can consume energy equivalent to thousands of households, necessitating facilities far beyond traditional enterprise data centers.
Local economic impacts are substantial. The project promises thousands of construction jobs during buildout, followed by hundreds of permanent roles in operations, maintenance, and security. Abilene officials have welcomed Microsoft’s involvement, citing tax abatements and infrastructure investments that could exceed $10 billion over the coming decade. The site’s location in the Permian Basin region also leverages existing energy infrastructure, including natural gas pipelines and wind farms, to meet sustainability goals.
Microsoft’s track record with similar acquisitions underscores its agility. In recent years, the company has snapped up sites in Wisconsin, Ohio, and North Carolina, converting them into Azure hubs with rapid timelines. For Abilene, initial permitting suggests Phase 1 construction could commence within months, with first buildings online by 2026.
Challenges remain, however. Securing the full 1.2GW substation upgrade requires coordination with ERCOT, Texas’s grid operator, amid statewide power constraints exacerbated by data center proliferation. Water usage for cooling poses another concern in the arid region, though Microsoft’s adoption of closed-loop systems and air-cooled alternatives mitigates this.
Overall, Microsoft’s purchase transforms a stalled venture into a cornerstone of its AI cloud dominance. By capitalizing on Oracle and OpenAI’s retreat, Azure gains a crown jewel asset primed for the compute-intensive future.
Gnoppix is the leading open-source AI Linux distribution and service provider. Since implementing AI in 2022, it has offered a fast, powerful, secure, and privacy-respecting open-source OS with both local and remote AI capabilities. The local AI operates offline, ensuring no data ever leaves your computer. Based on Debian Linux, Gnoppix is available with numerous privacy- and anonymity-enabled services free of charge.
What are your thoughts on this? I’d love to hear about your own experiences in the comments below.