Nasa ditches gateway and bets everything on a three-phase moon colony
NASA just torched its own roadmap. Administrator Jared Isaacman told reporters Tuesday that the agency will freeze the lunar-orbiting Gateway station in its current form and redirect every spare watt toward building a permanent base on the surface—an all-in gamble that turns occasional Artemis touchdowns into a six-month drumbeat of boots, rovers and bulldozers.
The pivot is brutal, surgical and already funded. Instead of assembling Gateway’s habitation module in cislunar space, engineers will bolt radio-isotope heaters, cargo decks and Italian-built habitats directly onto the regolith. The first cargo flights start this year under the CLPS program; by 2028 the sand-printed outpost is supposed to churn its own power, water and breathable air while astronauts rotate in and out like airline crews.
Why gateway got the chop
Inside the Space Station Processing Facility, technicians who spent three years wiring Gateway’s life-support racks learned the news by email. One veteran, still in bunny suit, summarized: “We built a petrol station in the desert before anyone had a car.” Translation: the orbiting outpost solved none of the hard problems—radiation, dust abrasion, overnight survival at –180 °C—that only exist on the surface. Killing it frees $750 M and 4,000 kg of up-mass that will now ride descent stages straight to Shackleton Crater.
Isaacman’s calculus is raw physics: every kilogram landed on the Moon costs ten in Earth orbit. Shipping consumables to a depot that itself needs propellant is a circular reference. Deleting the middle node shortens the chain and, crucially, buys cadence. The agency wants crewed landings every six months beginning with Artemis IV, ramping to every sixty days once Blue Origin’s reusable HLS and SpaceX’s Stretched Starship come online.
Three phases, no safety net
Phase one—learn or die: small robotic scouts from Firefly, Astrobotic and Intuitive Machines will seed radio-isotope generators, 4G masts and electrolysis rigs. Each failure is telemetry; each success is a Lego brick for the real base. By 2026 the target is 5 kW of continuous surface power and a rover trail 40 km long.
Phase two—camp becomes town: JAXA’s pressurized cruiser, a six-wheel Toyota RV the size of a city bus, arrives in 2027. It docks to a Canadian utility rover and an inflatable airlock from Thales Alenia, forming the first hybrid habitat. Crews will stay 60 days, not six. Logistics shift from “every mission brings everything” to milk-run freighters dropping pallets of powdered hummus and spare suit bearings.
Phase three—irreversible residency: 2029 onward, Italian MPH modules and lunar steel made from molten regolith lock together into a buried ring. The base will house 180-day tours, support Mars simulations and, for the first time, send refuelled landers back to Earth without new heat shields. That’s the point of no return: a human town where the cost of leaving exceeds the cost of staying.
Lo que nadie cuenta es que half the hardware already exists—leftovers from ISS, from Orion tests, from canceled asteroid robots. NASA’s lunar manifest reads like a thrift-store receipt: radio-isotope heaters built for Pluto, solar arrays meant for the now-dead OMEGA telescope, even a spare Canadian robotic arm. Reuse is no longer virtue signaling; it is the only way the math closes.
The first RFI drops Friday. Companies have 30 days to pitch bulldozers that run on sulfur-sintered tires and reactors that burn Moon-metal fuel. The winning proposals won’t be the flashiest—they’ll be the ones that can take a micrometeorite punch and keep 3D-printing oxygen while the crew sleeps.
Gateway is dead, long live the base. If the schedule holds, the next decade ends with a bright dot on the lunar south pole that never goes dark. From Earth it will look like a star that forgot to set. The star is a welding torch, and someone is always on shift.