-
Global gas turbine orders exceed 100 gigawatts, while manufacturers can produce an estimated 60–70 gigawatts annually.
-
US developers face average delivery waits of six years for new turbines, with some orders extending into 2030 and 2031.
-
AI infrastructure company Crusoe has ordered 13 aircraft-engine-derived turbines with a combined capacity of 650 megawatts.
-
Used equipment offers developers a faster route to electricity, although rising prices, maintenance risks and emissions complicate the economics.
November 09, (THEWILL) – An aircraft engine that has completed its flying career may soon find another job powering artificial intelligence.
Data centre developers are buying used gas turbines, reviving retired generating equipment and converting aircraft engines into electricity generators as the global AI boom stretches manufacturers’ production capacity beyond what they can deliver.
The scramble has created an unusual market. Equipment that power companies once retired or sold at a discount now attracts buyers eager to avoid waiting years for new machinery.
Reuters reported on Monday, November 9, that global gas turbine orders have exceeded 100 gigawatts, while manufacturers can currently produce an estimated 60 to 70 gigawatts annually, according to energy consultancy Wood Mackenzie.
The pressure is particularly severe in the United States, where a new turbine takes an average of six years to arrive. Some manufacturers are discussing deliveries for 2030 and 2031.
The shortage threatens to slow AI projects even after their developers have secured investment, computing equipment and suitable sites. Servers cannot operate without electricity, and a company that finishes constructing a data centre before securing enough power may struggle to put the investment to work.
When an Aircraft Engine Becomes a Power Plant
American energy company ProEnergy has developed one response to the shortage by rebuilding retired aircraft engines into stationary gas turbines.
Its PE6000 system starts with a General Electric CF6-80C2 engine core, a design used in commercial aircraft including the Boeing 747 and 767.
Engineers modify the equipment to drive a generator rather than propel an aircraft. Each completed unit can produce approximately 48 to 50 megawatts of electricity, depending on its configuration.
AI infrastructure developer Crusoe ordered 13 of these turbines in April, securing a combined capacity of 650 megawatts for its American data centre projects.
The order demonstrates how developers are moving beyond conventional electricity procurement. Rather than depend entirely on utility companies and grid connections, some are assembling generating capacity alongside their computing facilities.
ProEnergy has previously described its converted turbines as temporary sources of primary power that can operate during construction and the early years of a project.
Once a data centre secures a suitable grid connection, its operator may retain the turbines for backup electricity, redirect them to another project or sell them.
The approach makes commercial sense when developers face lengthy delays. A generating unit that arrives earlier can help a company begin operating expensive computing infrastructure instead of leaving it idle.
Aircraft engines are not the only alternative. Developers are also purchasing refurbished industrial turbines and smaller generating units, sometimes combining equipment from different manufacturers to reach the required capacity.
Those choices introduce practical problems. Engineers must establish the condition of older equipment, assess how much operating life remains and secure replacement parts. Systems originally built for different purposes may also need modifications before they can support continuous electricity generation.
The Price of Not Waiting
Demand has pushed prices higher across the second-hand turbine market.
Reuters cited equipment buyers who placed the cost of well-maintained used turbines at approximately $1.5 million to $1.7 million per megawatt, compared with $2 million to $2.5 million for new machines.
Refurbished units can cost around $1.3 million per megawatt, although prices vary with condition, capacity and the work required before installation.
One equipment dealer reported selling two used 14.5-megawatt turbines for $12.5 million. Comparable new equipment would have cost approximately $35 million and required a five-year wait.
The discount matters, but developers are paying for something beyond machinery. Earlier access to electricity can determine how soon they begin earning revenue from computing facilities.
The shortage has also opened opportunities for manufacturers offering alternative technology. Finland’s Wärtsilä reported approximately three gigawatts of orders from American data centres for power-generating gas engines since spring 2025.
Traditional utilities now compete with AI developers for some of the same equipment. That competition could make expansion more expensive for electricity suppliers serving households, factories and other businesses.
However, converted aircraft engines cannot solve the entire shortage.
Available second-hand equipment remains limited, while gas supply, environmental permits and grid connections can delay projects even after developers acquire turbines. Operating fossil-fuel generators also raises questions about emissions, efficiency and the long-term cost of producing electricity independently.
The global rush has implications for countries seeking to develop AI infrastructure of their own.
Nigeria, for example, wants to attract more investment into data centres and digital services, but electricity reliability already represents a major operating challenge. Competition for power equipment could add another cost to projects that need dependable generation or backup capacity.
There is no verified evidence that Nigerian data centre operators have joined the international market for retired aircraft engines. The developments elsewhere nevertheless show how important electricity procurement has become to the economics of AI investment.
Technology companies may continue increasing their spending on processors and servers, but the speed of expansion increasingly depends on equipment that has little to do with writing software or developing AI models.
Some of the world’s newest computing facilities are now competing for machinery built decades ago, including engines that once carried passengers across continents.
The race to develop more powerful artificial intelligence has made dependable electricity, and the ability to obtain it sooner than competitors, a valuable advantage.
Joy Onuorah is a business journalist and brand communications specialist covering financial markets, artificial intelligence, digital economy, and the ideas reshaping business across Africa and the global market. Beyond her reporting for TheWill, Joy uses brand strategy, storytelling, copywriting, and high-value SEO to help brands build lasting market authority.
