What Comes Next for Artemis?

With the successful completion of Artemis II and safe return of its crew to Earth, NASA sets its sights on the next Artemis mission, Artemis III, which NASA aims to launch in mid-2027. Originally planned as the first crewed lunar landing since 1972, the Artemis III mission was revised by NASA in February 2026 as part of broader changes to the Artemis program. The amended Artemis III mission will aim to test one or both of the Human Landing Systems (HLSs), conduct rendezvous operations between the landers and the Orion spacecraft, and test new spacesuits designed for lunar operations in low Earth orbit (LEO). The crewed lunar landing will now occur as part of the Artemis IV mission, scheduled for no earlier than 2028. NASA also canceled the Exploration Upper Stage (EUS), an upper-stage rocket under development for the Space Launch System (SLS), scrapped plans for building more powerful versions of the SLS rocket, and paused work on Gateway, a planned lunar space station. Instead of Gateway, NASA will build a base on the Moon’s surface near the lunar south pole.

Commendably, the new plan attempts to address a risk that few have emphasized—the risk that would have come from operating complex systems (i.e., Orion and the lunar lander) together for the first time, nearly 400,000 km from Earth, during a mission to land humans on the Moon for the first time in over 50 years. But the revisions also generate new risk that could ultimately impact the schedule and cost of the Artemis program. The revised program also does not address the risk that many have been talking about for years: Will the HLS be ready in time? Additionally, the new plan does not address concerns about the development of new lunar spacesuits for the Artemis astronauts.

Q1: What does NASA hope to achieve with these changes?

A1: According to the NASA administrator, the Artemis revisions will allow NASA to “move faster, eliminate delays” and achieve its objectives by “standardizing vehicle configuration, increasing flight rate.” NASA has drawn parallels between this new plan, which it calls a “logical, phased approach,” and NASA’s approach in the 1960s during the Apollo program. Under the old plan, the Lockheed Martin–built Orion spacecraft and HLS—either the lander developed by SpaceX or Blue Origin—would have conducted their first rendezvous and docking operations in orbit around the Moon as part of the lunar landing mission (i.e., Artemis III under the old plan). The new plan is intended to reduce the technical risks associated with these operations and with the Axiom-manufactured lunar spacesuits, which the company has announced will be tested in space in 2027, during a new mission conducted in low Earth orbit (i.e., Artemis III under the new plan), specifically designed to test out these systems working together in space.

By canceling the EUS and SLS upgrades, NASA aims to create one standardized configuration of the SLS rocket that would include the Boeing-built core stage, Northrop Grumman–built solid rocket boosters, and a to-be-developed variant of the Centaur V upper stage, which is built by United Launch Alliance (ULA) and currently used on the Vulcan rocket. The development of the EUS had been criticized for cost and schedule overruns. The modified Centaur V upper stage would replace the Boeing-built Interim Cryogenic Propulsion System (ICPS), the SLS upper stage flown on both Artemis I and Artemis II. The decision to award ULA with a contract to develop a modified Centaur V—not necessarily to replace the EUS, but to create a “standardized” SLS upper stage—was based on the assumption that, according to NASA, an “award to another source would cause unacceptable delays to current launch schedules.”

Canceling the EUS development also effectively doomed Gateway, as Gateway relied on the upgraded versions of the SLS rocket, including the EUS, for transportation of its components into lunar orbit. Therefore, it was not surprising that NASA, as part of its announcement of changes to the Artemis program, paused work on Gateway. NASA had already outlined a goal of establishing a nuclear-powered lunar base and had removed Gateway from the critical path for lunar missions. As part of the February 2026 announcement, the NASA administrator announced a pause on Gateway and plans to develop a new base on the Moon’s surface, stating it would cost $20 billion over the next seven years to build the initial elements of the base. For comparison, NASA has already awarded contracts valued at around $4.4 billion for work on Gateway. Additionally, international partners (i.e., the European Space Agency, Japan, Canada, and the United Arab Emirates) have likely collectively spent billions of dollars on and produced hardware for their contributions to Gateway.

Q2: Will the lunar landers be ready in time?

A2: In 2023, a U.S. Government Accountability Office (GAO) report raised concerns about the risk to the Artemis schedule from the development of SpaceX’s Starship HLS. In a 2026 report, NASA’s Office of the Inspector General raised additional concerns about the HLS development and highlighted the risk that the lander might not be ready for a 2028 lunar landing. One key observation is that SpaceX has neither tested in-space refueling nor sent the Starship HLS on an uncrewed test mission to the Moon—two key milestones in the development of an operational system. For the Starship HLS to work, SpaceX will have to conduct repeated on-orbit cryogenic fuel transfers from one Starship to another in orbit. Tests to demonstrate both of these capabilities had been planned for 2025, but were later moved to 2026. To date, SpaceX has yet to flight test Starship V3, the newest variant of Starship, which would underpin the Starship HLS. The first test of the Starship V3 has been delayed to at least May 2026.

In November 2025, NASA sought ideas from Space and Blue Origin, which had been on contract to design its own HLS, called the Blue Moon Mark 2, for future Artemis missions, on how to accelerate work on their respective lunar landers. In response, Blue Origin has accelerated work on its lander. The company has, for example, paused its sub-orbital New Shepard flights to focus resources on the Moon lander project. In early April 2026, Blue Origin announced that a Mark 1 cargo-only version of its HLS had successfully completed its most recent testing, which involved vacuum chamber tests at the Johnson Space Center.

Under the original Artemis plan, at least one of these two landers would have been needed in time for the planned 2028 crewed lunar landing. Under the new plan, one or both landers will be needed a year earlier, in 2027, to support the revised Artemis III mission in low Earth orbit. There is, however, nothing in the revised plan that specifically aims to accelerate development of the landers beyond what NASA has already done to achieve that goal. NASA will undoubtedly continue to put considerable pressure on both vendors, encouraging and assisting them in whatever way possible, so that they deliver as quickly as possible flight-ready hardware. But the risks associated with the development of the HLS landers remain as acute now as they did before NASA revamped the plan for Artemis.

Q3: What is the status of the Artemis space suits?

A3: As with the HLS lander, the GAO has raised concerns about the development of space suits for lunar missions, most recently in a 2025 report. In that report, the GAO expressed concern that Axiom’s space suits might not meet “certain NASA requirements” and that, after Collins ended its work to develop lunar space suits, NASA was stuck “having one space suit provider.” Axiom Space, the sole company now designing space suits for the Artemis program, was founded to build and operate a commercial space station. The destiny of this company is intrinsically tied to NASA plans for Commercial LEO Destinations (CLD), the program intended to replace the International Space Station (ISS). Since first announced by NASA about 10 years ago, CLD has faced delays and changes—it has also been chronically underfunded. NASA changed direction yet again on CLD as part of the same February 2026 announcement outline changes to Artemis, having signaled in August 2025 that it intended to accelerate work on CLD.

In March 2026, the president of a trade group representing companies involved in CLD work emphasized the need for stability and progress from NASA on its way forward for CLD. Meanwhile, NASA has yet to issue a final call for industry proposals. In addition to the lunar space suits and commercial space station, Axiom markets and sells commercial flights, operated by SpaceX, to the ISS and low Earth orbit. But fundamentally, for Axiom to survive as a business, it must secure funding from NASA for its commercial space station—this is the elephant in the middle of the room. Perhaps Axiom could get by for a time with only revenue from commercial astronaut missions. However, this would only be temporary, as the ISS will be retired in the coming years. Where will those commercial astronauts go without the ISS? Thus, the future of CLD is tied to the future of Artemis. Without proper funding and progress on CLD, leaving Axiom on a starvation diet, NASA risks jeopardizing the business viability of the only company building lunar space suits. Without these space suits, no American would have taken any steps on the Moon.

Q4: What are the new risks associated with the Artemis revisions?

A4: The revised plan introduces a number of new risks that could impact the schedule and cost of Artemis. First, NASA will need to stack and ready an SLS rocket for the revised Artemis III launch in less than a year. Boeing is already building the core stage for both Artemis III and Artemis IV and would likely have no problem delivering those components on time to meet this new schedule. The new risk arises from the plans for a new upper stage. The upper stage of the SLS, used for both Artemis missions to date, the ICPS, is needed to get the Orion spacecraft and its astronaut to the moon. The ICPS is not needed for operations in low Earth orbit. Therefore, there is no need to fly an ICPS for the revised Artemis III mission—and, besides, NASA only has one ICPS left, which presumably should be saved for the lunar landing mission. But NASA cannot simply stack the Orion spacecraft on top of the core stage; NASA would need to develop some type of filler that would go between the core stage and the Orion spacecraft. While it is tempting to write off such a system as trivial, it is still important. Designing a proper filler will be critical to ensuring the success of Artemis III. This will take time and money.

But no matter whether NASA uses the last ICPS for the revised Artemis III mission or saves it for Artemis IV, NASA still needs an upper stage for future Artemis missions. Ostensibly to save money and time, NASA has ditched the EUS in favor of the Centaur V. In its sole-source justification for awarding the contract to ULA, NASA states that the Centaur V was the only suitably available commercial option. The Centaur V was designed as the upper stage for ULA’s Vulcan rocket, allowing Vulcan to place uncrewed payloads into higher orbits and other trajectories beyond low Earth orbit. But for its planned use with the SLS, the actual version of the Centaur V that will fly with the SLS will be very different from the version used with the Vulcan rocket. While ULA can no doubt make those modifications, they carry risks of impacts to the Artemis schedule. Another question is how many modified Centaur V upper stages would be needed. Beyond Artemis V, does NASA plan to continue using the SLS or would it prefer to switch to a different vehicle, like Starship? If the plan is to switch to Starship, the “standardized” SLS configuration using the modified Centaur would fly, possibly only once.

Q5: Is the Centaur V ready—technically and procedurally—for human spaceflight?

A5: The Centaur V is not rated, from a safety and certification standpoint, to carry humans into space. And it would need such a certification to launch Artemis astronauts. Additionally, the engine would need significant design changes to work as a replacement for the ICPS. In particular, the Centaur V cannot hold the weight of the Orion spacecraft and would collapse in on itself like a soda can if used with its current specifications. While these changes and others that will likely be required may appear small, they still add unknowns to the schedule and cost for Artemis. Calling this variant commercial would be equivalent to calling the U.S. Air Force KC-46A tanker a commercial because it is a derivative of a Boeing 767 commercial jetliner. It should be recalled that the tanker development was plagued by cost overruns and delays. Whether a clean-sheet aircraft would have actually been quicker and cheaper is a question whose answer can never be known.

Finally, adding a new Artemis mission before the one carrying astronauts to the lunar surface adds another complication. NASA has only one place, one parking spot, in the giant Vehicle Assembly Building (VAB) at the Kennedy Space Center to stack SLS rockets. It takes a while to stack an SLS rocket. NASA began stacking the SLS used for Artemis II in March 2025. NASA also only has one suitable mobile launch platform on which the SLS sits before launch. As already noted, there are concerns that a lunar lander would not have been able to make NASA’s original timeline of landing astronauts on the moon in 2028. But now NASA is asking for one or both lenders in 2027. If no lander is available in mid-2027, presumably, NASA will delay Artemis III until one or both of those landers are ready. That means a fully assembled SLS will likely wait in the VAB, preventing NASA from beginning the process of stacking another SLS for the Artemis IV lunar landing mission. Every day that a lunar lander is not available for the revised Artemis III will, thus, add one day's delay to the lunar landing mission.

Q7: What comes next for Artemis as NASA works to manage both old and new risks?

A7: To recap its inarguably positive aspects, the revised Artemis plan acknowledges and attempts to buy down some risk that few were talking about. No longer is NASA planning to operate together for the first time critical hardware—the Orion spacecraft and lander—in lunar orbit during the high-stakes lunar landing mission. But taking action to address this risk creates new ones. NASA should be clear-eyed about these new risks and recognize that the new plan has just as much potential, but in different areas, to affect the planned 2028 lunar landing as it did the original plan. NASA should also continue efforts to buy down risks that carry over from the old to the new plan—the risk that a lunar lander and space suits will not be ready in time. With respect to the space suits, that means recognizing the linkage between CLD and Artemis that results from Axiom’s business case. For the lunar landers, that means taking proactive steps to minimize the impact of possible delays in delivery by SpaceX and Blue Origin of HLS for Artemis III to the timeline for Artemis IV. This might involve work now to allow for the simultaneous stacking of two SLS rockets at the Kennedy Space Center—a less-than-ideal outcome, but one that might allow Artemis IV to stay on track. To quote Proverbs, “do your best, prepare for the worst.”

Clayton Swope is the deputy director of the Aerospace Security Project and a senior fellow in the Defense and Security Department at the Center for Strategic and International Studies in Washington, D.C.