OpenAI's Stargate initiative has already surpassed its 2029 goal less than a year after announcing it. The company committed to securing 10GW of AI infrastructure in the United States by 2029, but has already exceeded that milestone, with more than 3GW added in the last 90 days alone. The demand for AI compute is accelerating faster than anticipated, pushing the company to expand capacity rapidly.

The Abilene, Texas facility serves as the flagship Stargate location, running on Oracle Cloud Infrastructure with NVIDIA GB200 systems. This is where GPT-5.5 was trained, demonstrating the site's capability to handle frontier AI models. OpenAI is building through partnerships rather than going it alone—collaborating with Oracle, NVIDIA, Vantage Data Centers, and others to share expertise across cloud infrastructure, chips, energy, and construction. The company emphasizes that no single entity can build this infrastructure alone; it requires coordination across the entire ecosystem.

The Abilene site uses closed-loop cooling systems that consume minimal water—roughly four average households' worth at full capacity. OpenAI is also investing in local communities through workforce development with building trades unions and educational donations like the one to Wisconsin's Port Washington-Saukville Education Foundation. This approach reflects a broader strategy: building AI infrastructure that creates tangible local benefits while meeting explosive demand for compute capacity.