- OpenAI’s ‘Guaranteed Capacity’ allows large enterprises to lock in up to three years of guaranteed computing
- In return, OpenAI earns predictable revenue to fund continued expansion of the data center.
- Stargate also introduced a new closed-loop water cooling system in a major leap in sustainability.
OpenAI, maker of ChatGPT, has announced a new “Guaranteed Capacity” plan for companies that want to block and reserve long-term access to the company’s IT infrastructure.
This strategic shift comes as AI companies face increased pressure from growing enterprise adoption of AI, which can lead to occasional downtime.
OpenAI recently revealed that around 40% of its $2 billion monthly revenue comes from enterprise customers, a figure it expects to continue to grow.
Guaranteed Capacity
The company recently revealed that it had 900 million weekly ChatGPT users, but before WSJ Reports indicated that OpenAI failed to meet internal goals in 2025, including the number of users.
That article also revealed possible tension between CDO Sarah Friar, who advocates financial discipline, and CEO Sam Altman, who takes a “buy everything” approach.
The Guaranteed Capacity program is designed specifically for large-scale applications and agent automation that require continuous, uninterrupted computing, rather than smaller businesses that want to avoid occasional interruptions.
Eligible businesses can request over 1 billion tokens per minute in terms of capacity, with pricing tied directly to annual spending levels across the one-, two-, and three-year plans.
“Customers can leverage this commitment across the OpenAI product portfolio,” the company said.
The company’s CEO Sam Altman said in an X post that the update comes in response to customer feedback and would be available “until we exhaust our current allocation for this program.”
“We will significantly expand our computing capacity,” says OpenAI
Altman also admitted that predictable revenue from large enterprise contracts and Guaranteed Capacity applicants would help the company plan, giving it access to more consistent cash to build future data centers.
“Our intention remains to build as much compute as quickly as we can,” Altman added.
The main theme of OpenAI is the Stargate project, which attracted collaboration from Microsoft, Oracle and others. When first announced in January 2025, the company was targeting 10 GW of AI infrastructure in the US by 2029, a figure it had reached just over a year later. In April, the company said it had also recently acquired another 3GW of capacity.
With plans to “significantly expand [its] computing capacity in the coming years,” OpenAI and its partners are already exploring future locations in the US, noting the strides data centers have made in terms of sustainability as well.
Each building at its flagship Abilene, Texas, campus requires the equivalent of two Olympic-sized pools of water, but instead of traditional evaporative cooling towers, it uses a closed-loop system where water recirculates through sealed pipes.
“Annual water use for the entire cooling system at full installation is expected to be comparable to that of a medium-sized office building, or approximately four average homes,” the company stated.
By offering enterprises a way to secure up to three years of dedicated computing, OpenAI is ultimately building predictability into their revenue to sustain ongoing data center buildout and further improve efficiency.
Follow TechRadar on Google News and add us as a preferred source to receive news, reviews and opinions from our experts in your feeds.
