MONTH 0 · OCTOBER 2026
The Plan
Galactic builds solar power stations in orbit and beams the power down to Earth. To get them there, we are building our own launch system.
Solar is the fastest-growing source of power in history, but it still makes only about 6% of the world’s electricity. On the ground, solar panels stop at night, lose output to clouds, and need a lot of land.
In high orbit, a solar panel is in sunlight almost all the time. A power station there can collect energy around the clock and send it to receivers on Earth as microwaves. None of this requires new physics. Solar cells date back to the 1950s, and NASA demonstrated wireless power transmission in 1975.
The hard part is launch. A useful space solar fleet weighs millions of tons. Rockets are too expensive, and fly too rarely, to put that much in orbit.
So we are building Bifrost, a launch loop: an electromagnetic track about 2,000 kilometers long, built over the ocean near the equator. It runs on electricity instead of fuel, and can send a payload toward orbit every few minutes. Our target cost is $3 per kilogram. Rockets charge about $2,500.
The plan rests on three bets:
- Clean power that runs 24/7 is the most valuable kind of energy, and space solar can provide it.
- Space solar beats solar on the ground once launch costs fall below $10 per kilogram.
- Rockets will never get that cheap. Only a launch system without rockets can.
Below is how we get there, one milestone at a time.
About 90% of a rocket’s mass is fuel, and the whole world managed 317 successful launches in 2025. A space solar fleet that delivers 1,000 GW needs about 5 million tons in orbit.
Launching that on Starship would cost about $3.75 trillion. On Falcon Heavy, about $12.5 trillion. Either way it would take more than 80 years. On Bifrost, it would cost about $15 billion and take about two and a half years.
MONTH 3 · PHASE I
Rotor Control
We show we can control Bifrost’s rotor in a tabletop test.
Bifrost is based on a design Keith Lofstrom published in 1985. A long iron belt, called the rotor, runs in a loop inside a tube. It moves fast enough that its momentum holds the raised track up. Magnets keep it centered in the tube and turn it around at each end.
Everything depends on keeping the rotor stable at high speed. That is the hardest engineering problem in the system, so we solve it first, on a test bench.
The physics has been public for 40 years. What was missing was a customer that needed millions of tons in orbit. Our own solar constellation is that customer, and orbital data centers are becoming another.
MONTH 12 · PHASE II
Sub-Scale Rotor
A smaller version of the rotor reaches full operating speed.
Next we build a smaller rotor and run it at the speed the full system needs. This tells us whether the design holds up at operating speed before we build it at full size.
MONTH 18 · PHASE III
Full-Scale Rotor
We commission the full-size rotor at our launch site.
Bifrost needs a 2,000-kilometer stretch of open ocean near the equator that is flat, straight, and safe from bad weather. Only a few places on Earth meet those requirements.
We plan to secure the rights to the best of these sites first. Once a loop is built on one, nobody else can use it. That is a large part of our competitive advantage.
MONTH 24 · PHASE IV
First Suborbital Payload
Bifrost launches its first payload on a suborbital path.
The first launch doesn’t go all the way to orbit. It tests acceleration, release and tracking with a real payload, which are the same steps an orbital launch uses.
MONTH 30 · PHASE V
First Orbital Payload
Bifrost puts its first payload in orbit. Launch now costs $3 per kilogram.
From here, Bifrost can send payloads to orbit every few minutes, day and night. Most of what it costs to run is the electricity.
Bifrost is also a business in its own right. The orbital launch industry is worth about $30 billion today. At full capacity, Bifrost could carry 3 million tons a year for satellite operators and orbital data centers. At an average price of $20 per kilogram, that is $60 billion a year in launch revenue.
THE NEXT 12 MONTHS
Deploying Prometheus
We launch our first solar constellation, Prometheus, in a single year.
Prometheus will deliver 50 GW of power to Earth. It is made of thin-film solar panels, microwave transmitters, lightweight support frames, and radiators to shed heat: about 206,000 tons in all. On the ground, ten receiving stations called rectennas take in 5 GW each.
Launching Prometheus on rockets would take about 2,000 flights and more than ten years, which is more than the world’s entire launch capacity. Bifrost does it in 12 months.
PROMETHEUS ONLINE
50 GW, Day and Night
Prometheus starts delivering power to Earth around the clock.
Because the stations are almost always in sunlight, they run about 95% of the time. That comes to about 416 billion kWh a year. Sold at $0.03 per kWh, it brings in about $12.5 billion a year. Sunlight costs nothing, so we expect margins of around 60%.
50 GW × 8,760 hours in a year × 95% uptime ≈ 416 billion kWh.
416 billion kWh × $0.03 ≈ $12.5 billion a year.
The same math gives about $125 billion a year at 500 GW and about $1.25 trillion at 5,000 GW. These are illustrative model figures, not guidance.
SCALING UP
Scaling to 500 GW
Each new gigawatt costs less than the one before it.
Bifrost runs on electricity, and by now Galactic produces a lot of it. The more power we put in orbit, the more launches we can run, and the cheaper each new station gets. Most of the cost is in the initial build, so nearly every new gigawatt goes straight to profit.
At 500 GW, power sales alone come to about $125 billion a year. Launch services and power sold directly to orbital data centers add to that.
WITHIN 10 YEARS
Matching Global Electricity
A single Bifrost loop can match the world’s electricity supply in under a decade.
The world generates about 3,700 GW of electricity today. With one Bifrost loop, we can put that much capacity in orbit in just under ten years. At 5,000 GW, our cost of power falls to $0.02 to $0.03 per kWh, cheaper than any power source on Earth.
At that scale, power, launch and orbital services together could bring in about $1.5 trillion a year.
EPILOGUE
Energy Too Cheap to Meter
Cheap, clean power without limits would change a great deal. Countries without reliable grids could get electricity without building one first. We could pull carbon out of the air and turn it into fuel, turn seawater into drinking water, and grow more food on the same land. It would also power industry in space.
ROCKETS VS. BIFROST
Same goal: a 1,000 GW solar fleet, about 5 million tons in orbit. Pick a launch system.
Fast and cheap enough to build a main power source for the whole planet, with a launch business on the side.
Lux Aeterna
CONTACT
We are assembling a lean, world-class team to attack this problem with absolute mission criticality. Join us.