The $4-billion launch Congress wouldn't kill

The White House's 2026 budget request put a number on the argument that has shadowed NASA's Moon program for a decade: $4 billion per launch for the Space Launch System, the < class="gloss-link" href="/topics/launch-vehicles" data-title="Concept" data-gloss="Rockets designed to carry payloads from Earth's surface into orbit or beyond, overcoming gravity and atmospheric drag through staged combustion.">rocket that lofts Orion toward the Moon, which the document calls 'grossly expensive and delayed' and 140% over budget. SpaceX flies its Starship for somewhere around $90 to $100 million a launch. The request proposed retiring SLS after the next mission and handing the Moon to Starship.

Artemis III, as now scoped, "isn't a Moon landing at all but a low-orbit rehearsal where astronauts dock with the lunar Starship for about a day and never go inside it."

Congress said no. Appropriators 'firmly reject the Administration's proposal to terminate SLS/Orion,' and a July 2025 law funded the rocket past the cutoff the budget had drawn. So the live picture in 2026 is messier than the cancellation headline: SLS still flies Artemis III in 2027, carrying its crew, and is now slated to retire only after Artemis V. Starship becomes the primary lunar architecture from Artemis IV onward, with NASA even studying whether Orion's crew could ride a Starship beyond low Earth orbit.

The detail that captures the awkward in-between: Artemis III, as now scoped, isn't a Moon landing at all but a low-orbit rehearsal where astronauts dock with the lunar Starship for about a day and never go inside it — the vehicle won't have life support or a finished interior yet. The cost gap is real and the direction is set toward Starship, but the rocket the budget tried to bury is the one knocking on the door.

The forty-fold per-launch gap is the whole case for the shift, and it isn't going away. What this fight settled is that a number that lopsided still can't override a program with contractors in dozens of states and a crew already named — the cheaper architecture wins on the spreadsheet years before it wins on the pad.