It started, like many things these days, by scrolling on X.
I read article after article about the power crisis hitting AI data centers: GPU racks were sitting idle, waiting not for chips, but for electricity. I texted Sam Altman, who confirmed that power was indeed a major constraint. I contacted our engineering team and discovered that they already had the outline of a plan to build a power turbine based on our Symphony supersonic engine.
After a few conversations, it became clear: AI didn’t just need more turbines, it needed a new, fundamentally better turbine. Symphony was the perfect new engine to accelerate AI in America. About three months later, we signed a deal for 1.21 gigawatts and began manufacturing the first turbine.
Today we are announcing Superpower, our new 42 megawatt natural gas turbine, along with a $300 million financing round and Crusoe as a launch customer. And above all: this marks a turning point. Boom is now on a self-funded path to Superpower and the Overture supersonic airliner.
I want to share the real story of how this happened and why supersonic technology is exactly what America’s energy crisis demands.
America doesn’t have 10 to 15 years to solve its electricity problem the old-fashioned way.
If you’ve been paying attention, you know that the United States is facing a real energy crisis. GPU racks are idling because they cannot be powered. Data centers compete for substations and interconnection queues. Meanwhile, China is expanding its power capacity at a wartime pace – coal, gas, nuclear, everything – while America struggles to get approval for a single transmission line.
The AI will not wait for us to repair the network. And the United States simply doesn’t have 10 to 15 years to build its electricity infrastructure the old-fashioned way.
Hyperscalers have already adopted their own plan B: behind-the-meter power plants. You saw XAI’s Colossus I and II in Memphis. OpenAI’s Stargate I in Abilene. These projects are powered by a set of aerodynamic natural gas turbines, which are essentially modified jet engines from the 1970s. There’s something brilliant about this approach: the transition from gigantic “frame” turbines to mid-sized “aero-derivative” turbine arrays reflects the computer industry’s shift from mainframes to blade servers.
The problem ? The “blade servers” of the energy world are old technology and they are exhausted. Because the most popular “aero-derivative” turbines are based on subsonic jet engines, they are most efficient when the outside air temperature is -50°F, as is the case when reaching Mach 0.8 at 30,000 feet. As outside temperatures rise, there is no choice but to slow down the engines, otherwise the turbine blades will literally melt. These turbines begin to lose power at around 50°F and by the time the temperature reaches 110°, as often happens in popular data centers like Texas, 30% of the production capacity is lost. Nonetheless, the major manufacturers are all behind schedule for the rest of the decade and none are building a next-generation, advanced-technology turbine.
A supersonic engine core makes the perfect power turbine
When we designed the Symphony motor for Overture, we built something no one else has built this century: an all-new large motor core optimized for continuous high-temperature operation.
A subsonic engine is designed for short bursts of power on takeoff. A supersonic engine is designed to work hard, continuously, under extreme thermal loads. Symphony was designed for Mach 1.7 at 60,000 feet, where effective temperatures reach 160°F, not the frigid -50°F conditions in which older subsonic engines operate.
This gives Superpower several critical advantages:
- Full power even with high ambient heat – Where existing turbines lose 20-30% at 110°F, Superpower maintains its total production of 42 MW without derating.
- Waterless operation – Existing turbines require huge amounts of water for cooling to avoid thermal derating in hot environments. This is not the case with superpowers. It remains at full power, without water.
- Cloud-native control and monitoring. Superpower inherits the telemetry and operations stack we created for XB-1. Each turbine streams real-time performance data, supports remote control and reports anomalies before customers even notice them.
Superpower and Symphony are based on virtually identical turboshaft engines. Both share the same core (HPC and HPT) and a lightly tuned bass coil. In place of the Symphony’s hollow-core titanium fan, Superpower adds two additional compressor stages as well as a three-stage free-running power turbine connected to a high-output generator on its own shaft. Additionally, the engines use slightly different fuel injectors, Symphony optimized for Jet A and Superpower for natural gas.
Increasing production supersonically: vertical integration
The legacy aerospace supply chain is crowded. When the mission is urgent and the supply chain is congested, you build the supply chain. The new Superpower Superfactory begins with a simple vision: raw materials on one side of the building, gigawatts of completed electric turbines on the other side. We have already started manufacturing the first parts and much of the production equipment needed to support 2 GW/year is on order. With this new funding, we are ready to accelerate even further.
If America wants to build at the speed AI requires, vertical integration is not optional. We have our own foundry and large-scale CNC machining capacity. We will have more to share about the Superpower Superfactory in early 2026.
Increasing production supersonically: vertical integration
The superpower is a bit like our Starlink moment, the most powerful accelerator we’ve ever had toward our core mission of making Earth dramatically more accessible.
The fastest way to get a Symphony engine certified for passenger transportation is to run its core for hundreds of thousands of hours in the real world, powering Earth’s most demanding AI data centers. Each hour of rotation of a Superpower turbine is an hour of validation for Symphony. Every gigawatt we deliver strengthens our vertical integration and manufacturing capacity. And with the profitability of Superpower funding the rest of the aircraft program, we achieved something rare in aerospace: creating a self-sustaining path to a new airliner.
Superpower also reminds me of what’s at the heart of Boom: a team willing to take on what others consider impossible, to do with a small team what large companies couldn’t even attempt.
