The 2013 file ended by watching the Shuttle retire and wondering aloud who would succeed it, and the answer thirteen years on is not a vehicle but a procurement model. NASA's astronauts fly on SpaceX, its science lands on commercial robots, its Moon lander is a Starship, and in June 2026 it hired Eric Schmidt's rocket company to fly to Mars, so the agency that built Saturn V now mostly writes cheques. Artemis II looped the Moon in April carrying the first crew out there since 1972, and it did so on an SLS the same government now wants to scrap - two days before it asked Congress to halve the science budget and cancel the return of samples from Mars. NASA is still capable of the sublime even as it turns into the general contractor of it. Kennedy chose to go because it was hard, and the open question is whether choosing who to pay to go still counts as choosing to go.
In April 2026, for the first time in over half a century, human beings flew around the Moon. Four of them, ten days, out past the far side where no crew had been since 1972, and home to a splashdown in the Pacific. Artemis II did nothing fancy - it didn't land, it looped the Moon and came back - and it was still the most extraordinary thing any space agency did all year, flown there on a rocket NASA is now preparing to throw away.
The file this archive opened on NASA in 2013 ended on a question it left hanging. It watched the Space Shuttle retire and noted, almost in passing, that several companies had thrown their hat in the ring to succeed it. Thirteen years on the answer is in, and it is stranger than anyone filing that report would have guessed: the Shuttle's successor was not a vehicle at all, but a way of doing business.
NASA used to build its own spacecraft. It built the Saturn V, it built the Shuttle, it owned the hardware end to end. The NASA of 2026 increasingly writes cheques instead. Its astronauts have ridden to the ISS on SpaceX's Crew Dragon since 2020. Its science reaches the Moon aboard commercial robots the companies build and fly themselves, NASA riding along as a paying passenger - Firefly's Blue Ghost set down cleanly in 2025, others have tipped over or crashed short, and the agency covers the fare and watches from the ground either way. The lander meant to carry the next astronauts down to the surface is a version of SpaceX's Starship. And in June 2026, NASA hired a rocket company owned by the former head of Google to fly a science mission to Mars. The agency that once announced it would choose to go now, more often, chooses who to pay to take it.
That shift is the whole story of the modern agency, and Artemis II flew straight into the hardest stretch of it. Two days after the crew launched, while they were still out near the Moon, the White House asked Congress to cut NASA's budget by nearly a quarter and its science programmes by almost half, terminating something like forty missions with the long-planned return of rock samples from Mars among them. The SLS rocket that had just carried four people around the Moon was written up in the same documents as grossly expensive and marked for retirement after one more flight. The new administrator, Jared Isaacman - a billionaire who chartered his own trips to orbit before the agency hired him to run it - backs the cuts, his wager being that NASA does more by building less and buying the rest.
The reason the Moon money survives while the science budget burns is a race, stated plainly: get American astronauts back onto the lunar surface before China lands its own, which Beijing intends to do before 2030. The landing itself, Artemis III, is now aimed at 2027, riding that Starship down. One front of the race has already quietly been conceded, because with Mars Sample Return cancelled, the first pieces of Mars carried back to Earth now look likely to arrive in a Chinese capsule rather than an American one.
The handovers run all the way down to the ground NASA stands on. The ISS comes out of the sky in 2030, dragged down by a SpaceX vehicle built for the job, and is meant to be replaced by privately-run stations the agency would rent rather than own. Except the tenants have not turned up. NASA itself now admits the orbital-tourism market it counted on to pay for those stations never materialised, which could leave it funding the "commercial" outpost on its own - the one place the buy-don't-build model keeps hitting a wall.
So the NASA in this drawer is two things at once, and they do not quite reconcile. It is still, on the evidence of a crew that looped the Moon in April, capable of the sublime, and it is at the same time becoming something the 1962 speech in Exhibit A never pictured - less the builder of the future than its general contractor, handing the actual flying to whoever bids lowest. Kennedy said the country chose to go to the Moon because it was hard. Sixty-four years later the open question is whether choosing who to pay to go still counts as choosing to go, and the next few years are the answer.