Chevron will supply natural gas-fired electricity to a major Microsoft Texas data center under a 20-year agreement that highlights how fast artificial intelligence is reshaping America’s energy map.
The project, known as Project Kilby, is planned for Pecos, West Texas, and is expected to become one of the largest co-located power and artificial intelligence infrastructure developments of its kind in the United States. Instead of relying only on the wider regional grid, the project is designed to place power generation close to Microsoft’s data center campus, giving the technology company a dedicated electricity source for its expanding cloud and AI operations.
The agreement brings together three forces that are now increasingly linked: Big Tech, natural gas and data center construction. Microsoft needs massive amounts of reliable power for AI workloads and cloud computing. Chevron has access to natural gas resources in the Permian Basin. West Texas has become one of the most important energy-producing regions in North America.
Project Kilby shows how the AI boom is no longer only a software or chip story. It is also an electricity story. Data centers require land, cooling systems, transmission infrastructure and, above all, steady power. As AI models grow larger and cloud services expand, energy supply has become a strategic priority for the world’s biggest technology companies.
What Is Project Kilby?
Project Kilby is a planned co-located power and data center project in West Texas. The facility is expected to provide dedicated natural gas-fired electricity to Microsoft’s data center campus for 20 years.
The project is expected to deliver first power by 2028 and eventually ramp up to approximately 2.67 gigawatts of generation capacity. That scale places it among the largest natural gas power projects tied directly to a data center campus in the United States.
A majority of the generation capacity is expected to come from GE Vernova gas turbines, with additional equipment supplied by Solar Turbines, a Caterpillar subsidiary. Chevron is developing the project with partners including Engine No. 1.
The project still requires a final investment decision and remaining approvals. Chevron expects that decision by the end of 2026. If completed as planned, Project Kilby would become a major example of how energy companies and technology companies are building direct partnerships to support AI infrastructure.
Why Microsoft Needs Dedicated Power
Microsoft is one of the world’s largest cloud computing companies, and its AI ambitions have significantly increased its need for electricity. Data centers that support AI tools, cloud platforms and large-scale computing systems consume huge amounts of power because they run thousands of servers, networking systems and cooling equipment around the clock.
AI workloads are especially power-intensive. Training and running advanced AI models requires high-performance chips that use more electricity than traditional computing infrastructure. As demand for AI services grows, Microsoft must secure enough electricity to keep its data centers operating reliably.
This is why dedicated power agreements are becoming more common. Instead of waiting for grid upgrades or competing with other industrial users for available electricity, major technology companies are looking for direct energy partnerships. These agreements can give them more certainty over power supply, timing and long-term infrastructure planning.
For Microsoft, Project Kilby is not just an energy deal. It is a strategic move to support future AI and cloud capacity in a market where electricity availability can determine how quickly new data centers come online.
Why Chevron Is Moving Into AI Power
For Chevron, the Microsoft agreement represents a major move beyond its traditional role as an oil and gas producer. The company is not only supplying fuel. It is helping develop power infrastructure for one of the most important growth sectors in the global economy.
Natural gas from the Permian Basin gives Chevron a strong advantage. West Texas is one of the most productive oil and gas regions in the world, and natural gas is often produced alongside oil. Turning that gas into electricity for data centers creates a new commercial opportunity.
This strategy also helps Chevron position itself in a changing energy market. Oil and gas companies are looking for ways to supply growing power demand, especially as AI, electrification and industrial expansion increase electricity consumption. Data centers offer long-term demand from financially strong customers, making them attractive partners for energy producers.
Project Kilby shows that the future of energy may not only be about selling fuel into traditional commodity markets. It may also be about building integrated power solutions for specific industrial and technology customers.
Why West Texas Matters
West Texas is central to this deal because it combines energy supply, land availability and industrial infrastructure. The Permian Basin is one of the most important oil and gas regions in the United States, giving Chevron access to nearby natural gas resources.
For data centers, location matters. Large campuses need enough land, power access, water planning, fiber connections and long-term operating stability. West Texas offers major energy advantages, but it also brings challenges, including grid constraints, water scarcity and infrastructure planning.
By co-locating power generation with the data center campus, Project Kilby is designed to reduce pressure on the broader regional grid. That does not remove every infrastructure issue, but it can help Microsoft secure a more direct power source while giving Chevron a clear demand center for gas-fired generation.
The project also reflects a wider trend: energy-producing regions are becoming attractive locations for AI infrastructure because electricity availability is now one of the biggest bottlenecks in data center growth.
How Natural Gas Fits Into the AI Boom
Natural gas is becoming a major part of the AI infrastructure conversation because it can provide dispatchable power. Unlike solar or wind, which depend on weather conditions, natural gas plants can generate electricity when needed. That reliability is important for data centers, which operate continuously.
AI data centers cannot easily pause operations when power supply is tight. They need stable electricity to run servers, cooling systems and network equipment. Natural gas can help meet that need, especially in regions where grid capacity is limited or new transmission lines would take years to build.
However, natural gas also brings environmental questions. Gas-fired power produces emissions, and technology companies face pressure to reduce their carbon footprints. Microsoft has made major climate commitments, so any natural gas-powered data center project will attract scrutiny from investors, regulators and sustainability advocates.
That creates a balancing act. Microsoft needs reliable power quickly. Chevron can provide fuel and generation. But the long-term debate will focus on whether projects like Kilby can align with corporate climate goals, future clean energy additions and broader decarbonization plans.
The Scale of the Project
Project Kilby is expected to ramp up to about 2.67 gigawatts of generation capacity. That is an enormous amount of electricity for a dedicated data center-linked power development.
To understand the scale, many traditional data centers operate at a fraction of that level. AI campuses, however, are changing the numbers. As companies build larger clusters of servers and AI chips, data center power demand is moving from hundreds of megawatts toward gigawatt-scale campuses.
This is why Project Kilby is important. It shows that AI infrastructure is entering an industrial phase. The largest projects now resemble major energy and manufacturing developments, not simply technology facilities.
The scale also explains why Chevron, GE Vernova, Caterpillar and Engine No. 1 are involved. A project of this size requires energy supply, turbines, capital planning, engineering, long-term contracts and regulatory approvals. It is not a small corporate power purchase agreement. It is a major infrastructure development.
Jobs and Economic Impact
Project Kilby is expected to support thousands of construction jobs and hundreds of long-term operational roles. Microsoft has said the wider data center investment will span several years, creating work during the construction phase and permanent employment after operations begin.
The project could also generate significant state and local tax revenue. Large data center and power infrastructure developments often affect local governments through property taxes, utility arrangements, roads, public services and economic activity around construction.
For West Texas communities, the economic impact could be meaningful. Construction workers, engineers, contractors, suppliers, equipment companies and service businesses may benefit from the project. Long-term operations could also create technical jobs tied to power generation, data center maintenance and site management.
At the same time, large projects can create local pressure. Communities may ask questions about water use, land use, traffic, housing demand and environmental impacts. Those issues are likely to remain part of the public conversation as the project moves toward approval and construction.
The Grid Question
One of the most important parts of Project Kilby is the decision to co-locate power generation near the data center. This approach is designed to provide Microsoft with dedicated electricity while reducing dependence on the wider grid.
That matters because data center growth is putting pressure on electricity systems across the United States. Grid operators are facing rising demand from AI, manufacturing, electrification and population growth. In many regions, new data centers cannot connect quickly because transmission capacity is limited.
Co-located generation can help solve part of that problem by bringing power directly to the site. It may reduce the need for long transmission upgrades and give the customer more certainty. But it also raises regulatory and planning questions about how private power facilities interact with public electricity systems.
As AI demand grows, more companies may follow this model. Large technology firms may increasingly build or contract dedicated energy sources instead of waiting for traditional grid expansion.
What This Means for the Data Center Industry
The Chevron-Microsoft agreement could become a model for future AI data center projects. It shows that hyperscale technology companies are willing to secure long-term power directly from major energy producers.
This could change the way data centers are financed and built. In the past, developers often focused first on land, fiber access and grid interconnection. Now, power supply is becoming the most important factor. A site with reliable electricity may be more valuable than a site in a traditional technology hub.
Energy companies may also become more active players in the data center market. Instead of selling natural gas only through commodity channels, they can develop power projects tied to long-term customers. This creates more predictable revenue and gives them exposure to AI-driven electricity growth.
The data center industry is therefore moving closer to the energy industry. Project Kilby is one of the clearest examples of that convergence.
The Bigger AI Energy Race
Project Kilby arrives as Microsoft, Amazon, Google, Meta and other technology companies race to secure enough power for AI growth. The competition is no longer only about chips, software engineers or cloud customers. It is also about energy access.
Some companies are signing renewable energy deals. Others are exploring nuclear power, battery storage, geothermal energy and natural gas. The mix depends on location, timing, cost, reliability and corporate climate goals.
Natural gas has become attractive because it can be built faster than many nuclear projects and can provide steady power when renewable generation is not available. But it also faces criticism because of emissions. That tension will shape the next phase of AI infrastructure development.
The winners in AI may be the companies that secure reliable, affordable and scalable electricity before their competitors. That is why Microsoft’s agreement with Chevron matters beyond one West Texas site.
Risks and Questions Ahead
Despite its scale, Project Kilby is not yet fully complete. The project still requires final investment approval and other conditions before construction and power delivery can move forward as planned.
Several questions remain. Will the project stay on schedule for first power in 2028? Will costs rise? How will regulators respond? How will water use be managed in West Texas? How will Microsoft balance gas-fired power with its climate commitments? Will future phases include battery storage, solar or carbon management?
These questions do not cancel the importance of the deal, but they show why the project will be closely watched. Large energy infrastructure developments can face delays, permitting issues and local concerns.
For investors and industry watchers, the key issue is whether Chevron and Microsoft can turn the agreement into a working model that other companies may copy.
Conclusion
Chevron’s plan to power Microsoft’s West Texas data center with natural gas marks a major moment in the relationship between energy and artificial intelligence. Project Kilby is more than a power plant deal. It is a sign that AI infrastructure now requires energy planning on an industrial scale.
Microsoft needs reliable electricity to support cloud and AI growth. Chevron has access to natural gas and the ability to help develop large-scale power infrastructure. West Texas offers the energy resources and land needed for a project of this size.
If completed as planned, Project Kilby could become one of the largest co-located natural gas power and AI data center developments in the United States. It could also become a model for how technology companies secure power in an era when electricity is becoming one of the most important resources in the AI economy.
The deal shows a clear direction for the market: the future of artificial intelligence will depend not only on chips and data, but also on power.
Read Also: Paramount-Warner Merger Raises Local Jobs Concerns
