From Flags to Factories: How the Race to the Moon Became a Race to Stay

On April 6, 2026, Reid Wiseman, Victor Glover, Christina Koch, and Jeremy Hansen flew 252,760 miles from Earth — farther from home than any humans had travelled since Apollo 13 in 1970. Their spacecraft, Orion, came within 4,070 miles of the lunar surface on the Artemis II mission, a ten-day crewed flyby that produced more than 7,000 photographs of the Moon and set a new distance record that the crew did not particularly want to think too hard about.

It was the most significant human spaceflight achievement in more than half a century. It was also, in the context of what is being built around it, a relatively modest step in a programme that has stopped thinking about the Moon as a destination and started thinking about it as a location.

The architecture is shifting

Three weeks before Artemis II launched, NASA made a decision that quietly rewrote the programme's logic. On March 24, the agency announced it was pausing work on the Lunar Gateway — the orbital station that was supposed to serve as a staging point for surface missions — and redirecting $20 billion over seven years toward a lunar surface base instead. The Gateway hardware may be repurposed. International partners, including the European Space Agency, were informed and are evaluating their response.

The Artemis III mission, which had been planned as the first crewed lunar landing, has been redesigned as an Earth-orbit rendezvous and docking test with SpaceX's Starship Human Landing System and Blue Origin's Blue Moon lander — not a Moon landing. Artemis IV, now targeting early 2028, will be the first crewed landing since Apollo 17 in December 1972. Beginning with Artemis V, NASA plans to start building permanent surface infrastructure with yearly missions.

The logic behind the pivot from a Gateway station to a surface base is mostly economic. Every kilogram of life support, power, water, and construction material launched from Earth costs roughly $1 million to deliver to the lunar surface. A self-sustaining base that extracts water from polar ice, produces oxygen from regolith, and eventually manufactures its own propellant changes that equation fundamentally. The Gateway approach deferred that challenge. The surface base approach makes it the central one.

Commercial landers are flying — with mixed results

The Commercial Lunar Payload Services programme, which NASA uses to deliver scientific instruments and technology demonstrations to the surface, has now run enough missions to produce a scorecard. The first attempt — Astrobotic's Peregrine in January 2024 — failed after a propellant leak. Intuitive Machines' Odysseus landed in February 2024 but tipped on a rock, limiting its operational life to six days. Two missions launched together in January 2025: Firefly Aerospace's Blue Ghost landed successfully at Mare Crisium on March 2, operating for a full lunar day and returning data from ten NASA payloads including a technology demonstration that used compressed gas to collect regolith samples from the surface. The ispace RESILIENCE mission, launched on the same Falcon 9, failed during its landing attempt in June 2025 when rangefinder delays caused insufficient deceleration.

The next wave arrives in the second half of 2026: Astrobotic's Griffin, Intuitive Machines' IM-3, a second Firefly Blue Ghost, and Blue Origin's first Pathfinder lander. China's Chang'e 7 is scheduled for August 2026, targeting the lunar south pole with a comprehensive environmental survey mission.

The pattern matters. Commercial landers are flying on a cadence measured in months, not years. The hardware is improving and the failure modes are being documented and corrected. Firefly's Blue Ghost was the first fully successful CLPS landing on a $101.5 million budget — a fraction of the cost of a traditional NASA science mission.

The resource case

The economic case for a permanent lunar presence rests on two resources: water ice and position.

Water ice exists in permanently shadowed craters near both poles in estimated quantities ranging from 300 million to over one billion metric tons, depending on measurement methodology. The Cabeus crater alone may hold 11 million tons. Ice can be split into hydrogen and oxygen, producing rocket propellant directly on the lunar surface. A "gas station" at the Moon's south pole would dramatically reduce the mass that must be launched from Earth for deep space missions — including eventual missions to Mars. Water is also essential for life support and radiation shielding, which the lunar surface provides inadequately on its own.

The second resource is positional. Low lunar orbit requires roughly 20% less delta-v to reach than Earth orbit from the lunar surface, making the Moon a logical staging point for the inner solar system. The presence of helium-3 — deposited by solar wind in estimated quantities exceeding one million metric tons — represents a potential fusion fuel resource worth roughly $4 billion per metric ton at current projections, though commercial fusion reactors capable of using it do not yet exist.

DARPA's LunA-10 programme, which contracted 14 companies including Blue Origin, SpaceX, Nokia, ICON, Firefly, and Sierra Space in December 2023 to study integrated commercial lunar infrastructure, identified five pillars that must develop in parallel: power (including nuclear fission for shadowed areas), pressurised habitation, in-situ resource utilisation, communications and navigation (the Moon has no GPS), and surface mobility. A $127 billion lunar economy is projected by PwC by 2050, though that figure depends on ISRU becoming viable, fusion energy advancing, and launch costs continuing to fall.

China is not waiting

China's Chang'e 6 returned 1.935 kilograms of samples from the Moon's far side in June 2024 — the first far-side lunar samples ever retrieved, from the oldest and deepest impact basin on the Moon. The data is still being analysed. Chang'e 7 follows in August 2026. Chang'e 8, planned for 2028, will test in-situ resource utilisation directly.

The International Lunar Research Station, jointly led by China and Russia with 12 participating nations including South Africa, Pakistan, and Egypt, is in its second phase through 2035 — construction of a basic station at the lunar south pole. Full operations and crewed missions are targeted for the phase beginning in 2036. Whatever one thinks of the geopolitical alignments involved, the ILRS represents a parallel permanent-presence programme that will be operating on a similar timeline to NASA's Artemis base and drawing on many of the same polar ice resources.

The infrastructure that actually makes this work

Permanent lunar presence requires solving problems that have no elegant solutions: regolith is electrostatically charged and extraordinarily abrasive, destroying seals and mechanisms with sustained exposure. Temperatures swing 250 degrees Celsius between lunar day and night. Radiation exposure is two to three times Earth surface levels with no magnetic field protection. A round trip for any resupply mission takes days minimum. Medical emergencies must be managed on-site for weeks before evacuation is possible.

The engineering responses to these challenges — nuclear fission power for the polar darkness, 3D-printed regolith structures for radiation shielding, ISRU oxygen extraction reducing life-support resupply weight — are being tested on CLPS missions right now. Nokia's 4G lunar surface network is being designed under the LunA-10 framework. Intuitive Machines has a NASA contract for south pole communications relay satellites.

None of this produces a permanent base by 2030. But the trajectory of missions, funding commitments, and infrastructure investment in 2025 and 2026 suggests that the question has shifted from whether humans will have a sustained presence on the Moon to when the infrastructure to support it becomes technically and economically viable. The $20 billion redirected from Gateway toward a surface base is the clearest institutional signal yet that NASA has decided the answer is worth committing to.