NASA Administrator Jared Isaacman laid out a $20 billion, seven-year plan March 24 to build a permanent human outpost at the moon's south pole — the most concrete U.S. commitment to sustained lunar presence since the Apollo era. The plan has three phases: robotic deliveries and terrain vehicle testing through 2028, early habitable infrastructure with a JAXA pressurized rover, then permanent human occupation requiring lander capabilities that do not yet reliably exist. It also formally pauses the Lunar Gateway station in its current form and redirects its hardware. And it comes with an add-on that has nothing to do with the moon at all: a nuclear-electric spacecraft launching to Mars in December 2028.
That spacecraft is Space Reactor-1 Freedom, and it is the part of the announcement with the most verifiable engineering detail. SR-1 Freedom is the first nuclear-electric interplanetary spacecraft — not a nuclear thermal rocket expelling hot nuclear gas through a nozzle, but a fission reactor generating electricity that powers xenon ion thrusters. It uses a 20-plus kilowatt fission reactor fueled by High-Assay Low-Enriched Uranium and Uranium Dioxide, a Boron Carbide radiation shield, and an advanced closed Brayton cycle power conversion system. Less than 48 hours after launch, the reactor activates. About one year later, the spacecraft arrives near Mars.
The launch vehicle is a Falcon Heavy. The propulsion and power bus is the Power and Propulsion Element — originally built for the lunar Gateway, then redirected to an asteroid redirect mission that was canceled, then reassigned back to Gateway, and now pointed at Mars. It already exists and has been powered up. Its mass is roughly 5,000 kilograms; Falcon Heavy can deliver 16,800 kilograms to Mars. That math works. The Skyfall payload: three Ingenuity-class helicopters, deployable mid-air after atmospheric entry, equipped with ground-penetrating radar to map subsurface water — reconnaissance for future human landing sites and in-situ resource utilization.
The moon base architecture depends heavily on landers that are behind schedule. The NASA Inspector General noted this month that SpaceX, awarded the astronaut lander contract in 2021, is two years behind schedule, per Reuters. Both SpaceX Starship and Blue Origin Blue Moon face complex engineering challenges before human flight. NASA's response, according to Reuters: whichever lander is ready first gets the mission, regardless of pre-assigned order. That is not a dual-provider strategy — it is a lottery with a moon landing attached.
The Gateway station was designed partly because Orion's service module engine lacks the thrust to get into and out of low-lunar orbit on its own. Without the Gateway, NASA is asking contractors to solve that orbital geometry problem by other means. No alternative has been announced. This is an open engineering question, not a solved one. The Gateway hardware was largely built by Northrop Grumman and Intuitive Machines subsidiary Lanteris Space Systems, per Reuters. International partners — ESA, JAXA, CSA — face uncertainty about their roles in the revised architecture.
NASA is targeting up to 30 robotic landings under the Commercial Lunar Payload Services program starting in 2027, and crewed landings every six months once the architecture matures, per SpaceFlight Now. That cadence assumes both landers work, which is the assumption least supported by the available evidence.
On budget, Isaacman's framing was unusually direct. "NASA does not necessarily have a top-line problem," he said, per CBS News. "We get a lot of resources. We may not always allocate them that efficiently." The Planetary Society estimates NASA will have spent approximately $107 billion on return-to-moon programs through 2026 in inflation-adjusted dollars — covering Constellation, the Obama-era asteroid redirect, and Artemis to date. The $20 billion figure is new incremental authorization.
China is targeting a crewed lunar landing around 2030, per Reuters. Isaacman put it plainly: "The clock is running in this great-power competition, and success or failure will be measured in months, not years."
The SR-1 Freedom development timeline is aggressive by any measure. Full vehicle goes to the launch site by October 2028, for a December launch — roughly 21 months of fabrication, integration, and testing for a spacecraft featuring a fission reactor, closed Brayton cycle power conversion, and xenon electric propulsion that has not flown this configuration together. Artemis II is eight days away as of the announcement. The architecture behind it is transitional: SLS is the bridge, not the destination.
What to watch: whether either lander reaches a flight-ready state before the 2028 crewed lunar target. Whether the orbital architecture question gets an answer before Phase Two infrastructure deliveries begin. And whether SR-1 Freedom ships in December 2028 — the first real demonstration that a closed Brayton cycle + fission reactor + ion thruster combination works beyond Earth orbit.
