NASA is no longer planning to land astronauts on the moon with Artemis 3. In an announcement following the successful Artemis 2 lunar flyby, acting NASA Administrator Sean Duffy confirmed that the agency has redesigned the mission entirely, replacing what was supposed to be the program’s first crewed landing with an Earth-orbit rendezvous and docking demonstration between Orion and the lunar landers built by SpaceX and Blue Origin. The reason is simple: neither lander will be ready in time.
The slip pushes the actual boots-on-regolith moment to Artemis 4 and Artemis 5, both scheduled for 2028. Whether that schedule survives contact with reality depends almost entirely on hardware that has never flown the way it now needs to — two private companies attempting to compress years of untested space hardware development into roughly 30 months.
What Artemis 2 Actually Bought NASA
The mission was, by any honest measure, a triumph. The crew flew a free-return trajectory around the moon, the first crewed lunar mission since Apollo 17 in 1972. Along the way they broke Apollo 13’s distance record, reaching a record distance from Earth.
The mission also did something programs of this scale rarely do: it landed emotionally. The crew spoke about the experience without polish. As BBC science correspondent Pallab Ghosh noted, the astronauts presented themselves in an unusually authentic way that resonated with the public — a small thing that translated into unusual public investment in what comes next.
Public investment matters. It is what makes multi-billion-dollar line items survive budget negotiations.
The Hardware That Doesn’t Yet Exist in Working Form
The reason Artemis 3 is no longer a landing mission is straightforward: neither lander is ready, and neither will plausibly be ready by late 2027. SpaceX’s Starship Human Landing System is still working through its V3 design, the version on which the lander variant is based, and has not yet flown. Blue Origin’s Blue Moon Mark-1 has completed ground testing and is expected to fly an uncrewed mission, but the human-rated Mark-2 vehicle is further out.
The capabilities both vehicles must demonstrate before they can carry astronauts have, in some cases, never been performed in space at all. Cryogenic propellant transfer between two vehicles in orbit — required for Starship to top off in low Earth orbit before heading to the moon — is the most prominent example. Uncrewed precision lunar touchdowns are another. Crew-rated life support systems for the lunar transit and surface stay are a third. None of these have flight heritage.
What NASA Is Actually Asking For
NASA has framed the redesigned Artemis 3 as a forcing function. Rather than wait until one lander is fully ready and risk a multi-year gap, NASA flies Orion to a high Earth orbit, both landers meet it there, and the agency gets real flight data on the docking interfaces, the propellant systems and the autonomous rendezvous performance — without betting four lives on any of it. If something fails, no one dies and the schedule absorbs the hit.
The budget request backs this up with substantial funding allocated to Human Landing System contracts, split between the two providers. That is a serious number, but it is not enormous given what it has to buy.
The Cadence Problem
The deeper issue facing Artemis is rhythm. Artemis 1 flew uncrewed in late 2022. That is a multi-year gap between missions for a program whose entire architecture assumes a sustainable presence on and around the moon.
NASA aims to compress future gaps significantly. To get from late 2027 to a landing in 2028 — and then a second landing in the same year — the agency needs to fly Orion three times in roughly 14 months while simultaneously certifying two new human-rated lunar landers. Apollo did something comparable, but Apollo had a workforce, a budget and a political consensus that Artemis does not.
SLS production cadence is part of this. So is Orion heat shield qualification, which dogged the program after Artemis 1. So is ground infrastructure, including the long-troubled Mobile Launcher 2 program that has cost the agency billions and is still not finished.
The Political Clock
Political pressure to achieve a landing during the current administration’s term creates both opportunities and risks for engineers. It tends to unlock budget and clear bureaucratic obstacles. It also tends to push schedules past the point where the people doing the work can honestly say the hardware is ready. The Apollo 1 fire, Challenger and Columbia all carried, in their post-mortems, some version of this same dynamic.
NASA leadership has been careful in its public language, framing the late 2027 rendezvous as a test, not a stunt, without making absolute promises about landing dates. That is a meaningfully different approach.
What to Watch in the Next Eighteen Months
Three milestones will tell us most of what we need to know about whether 2028 is real.
First, Starship V3’s debut and its subsequent demonstration of orbital propellant transfer. Without that, the SpaceX HLS architecture does not work. Recent V3 ground testing has been encouraging, but ground tests are not orbital tests.
Second, Blue Moon Mark-1’s uncrewed lunar landing attempt. A successful soft landing would give Blue Origin real flight data and put it on a credible path toward the human-rated Mark-2 vehicle. A failure would not kill the program, but it would compress an already tight schedule.
Third, whether NASA’s stated cadence target survives Artemis 3 preparation. If the agency cannot turn Orion around quickly, the 2028 landing dates become arithmetic fiction regardless of how the landers perform.
The honest answer to whether NASA can land astronauts on the moon in 2028 is that the agency has bought itself a credible plan, picked the right intermediate test, and put the burden squarely on two contractors who have a lot to prove. That is not the same as saying it will happen. It is saying the path is no longer obviously impossible, which, given where Artemis was a year ago, counts as progress.
Photo by T Leish on Pexels
