The Moon rocket flew a flawless mission in April. The budget retires it; the law forbids retiring it yet; nobody priced a mission.

The finding, in one paragraph: America's return to the Moon is being run under three money documents that cannot all be describing the same program.▸ The Inspector General's ledger says each Space Launch System/Orion flight costs $4.1 billion to produce and operate and that NASA "lacks a comprehensive and accurate cost estimate" for the enterprise it has spent $93 billion on. The reconciliation law says the rocket must be bought for two more missions, with minimum annual obligations written into statute — spending floors, for hardware, in a tax bill. And the fiscal 2027 budget request, published days after the rocket's first crewed flight succeeded, funds its commercial replacement "for Artemis VI and beyond" — while still giving the retiring vehicle six dollars for every dollar of its successor.

Five documents, read directly.▸ The cost record is NASA OIG's NASA's Management of the Artemis Missions (IG-22-003) — the $4.1 billion per-launch and $93 billion figures, and the finding that the agency's own estimate "excludes $25 billion for key activities." The statute is Public Law 119-21's NASA title, read verbatim on govinfo — roughly $10 billion of supplemental exploration funding with obligation floors. The plan is NASA's FY2027 Budget Request summary (April 2026). The event is Artemis II itself, per NASA's mission record: launch April 1, 2026, four astronauts, splashdown April 10.

The Inspector General's cost anatomy explains the $4.1 billion: the programs run on "sole-source, cost-plus contracts," and — except for the Orion capsule — "all components are expendable and 'single use' unlike emerging commercial space flight systems." Every launch throws away the rocket.▸ The OIG's warning, written before Artemis I flew, was that "without capturing, accurately reporting, and reducing the cost of future SLS/Orion missions, the Agency will face significant challenges to sustaining its Artemis program in its current configuration." Five years later the program is being unsustained on exactly that logic — but on a schedule set by statute, not by the cost data, which still does not exist in per-mission form.

Follow the sequence.▸ The FY2026 request proposed ending SLS and Orion after Artemis III, calling the rocket "grossly expensive and delayed." Congress answered in the July 2025 reconciliation law with appropriated floors: $4,100,000,000 for SLS for "Artemis Missions IV and V," "of which not less than $1,025,000,000 shall be obligated for each of fiscal years 2026, 2027, 2028, and 2029," plus Gateway floors of $750 million a year and station and infrastructure money — mandatory minimum spending, immune to the appropriations cycle. The FY2027 request absorbs the defeat and reschedules it: SLS and Orion fly Artemis III, IV, and V with the statutory money, and the new CoMMIT line — $744 million — buys "commercial capabilities to launch and return crews for Artemis VI and beyond." Retirement moved from III to V; the mandate and the termination now coexist in the same budget table.

Each document undermines another's premise.▸ The budget's case for retirement is cost — but the cost case rests on the OIG's $4.1 billion figure, because NASA, per the same OIG report, "lacks a comprehensive and accurate cost estimate" and used internal numbers that excluded $25 billion of Artemis activities. The statute's case for continuation is schedule certainty — but it guarantees inputs (dollars obligated per year), not launches, for a program whose first crewed flight slipped years. And the plan's case for the commercial hand-off assumes vehicles that do not yet carry crews beyond low Earth orbit will be ready when the mandated missions run out — a bet the OIG flagged in a different form when NASA selected "a single provider" for its lunar lander "after receiving $2.5 billion less than requested." The one undisputed data point is the newest: Artemis II worked. The system that produced a flawless crewed lunar flyby is the same one no document can price, retire, or commit to.

Artemis III — the landing — is the program's next public test, with the FY2027 request funding it on SLS/Orion.▸ The statutory floors run through FY2029 regardless. The CoMMIT procurement will show, in its solicitations, what "commercial crew to the Moon" costs when priced competitively — the first market comparison the program will ever have. And the appropriations committees, who have protected this rocket through three administrations, get the FY2027 request this summer; their answer, not the request, is the actual policy.

All figures are from the documents cited in-line, each read directly. The $4.1 billion per-launch figure is the OIG's projection for Artemis I–IV production and operations; NASA has stated its affordability initiatives aim to reduce it. Artemis II mission facts per NASA's mission pages.