A newly released NASA inspection report has exposed a deepening disagreement between the space agency and SpaceX over a fundamental design choice for the Starship lunar lander: whether astronauts should be able to manually pilot the vehicle as it descends to the Moon.
According to the Office of Inspector General audit released Tuesday, NASA's tracking shows SpaceX's proposed approach to manual control has grown worse over time, not better. The report states that "despite the provider's stated acknowledgment and commitment" to meeting NASA's manual control requirement, "NASA's tracking of SpaceX's manual control risk indicates a worsening trend."
The distinction between automated and manual landing is not merely academic. During every Apollo crewed lunar landing in the 1960s and 1970s, astronauts relied on backup manual controls to execute safe touchdowns. The inspector general's report notes this six-decade track record, pointing out that today's flight software, while far more sophisticated, cannot guarantee the lander will be safe without human oversight.
The tension reflects a genuine technical disagreement. SpaceX argues that its landing automation is robust and requires minimal astronaut intervention. NASA, however, views manual control as essential to crew survival. The report emphasises that Starship lacks the operational heritage that would justify flying purely on automation: "Starship will not have the same level of proven flight heritage in the actual operating environment for its crewed lunar missions," it states, contrasting the spacecraft's limited testing with the extensive flight data from crewed cargo missions to the International Space Station.
The issue carries real consequences. As the two sides approach the Critical Design Review, a key decision point that will clear the way for construction, an unresolved manual control requirement threatens to leave automation as the lander's only method. That outcome would represent a departure from the safety principle that has guided human spaceflight for decades: the presence of a human override.
NASA and SpaceX have faced this tension before. A decade ago, they clashed over Crew Dragon controls. SpaceX initially wanted touchscreens only, with limited astronaut commands available. NASA pushed back, advocating for systems equivalent to traditional spacecraft controls. The compromise, brokered with help from former NASA astronaut Garret Reisman (who was then at SpaceX), allowed crews to manually fly the capsule using controls embedded in touchscreens. That precedent suggests both sides can find common ground, but it also shows how hard these negotiations can be.
For Blue Origin, the situation remains different. The company has not yet submitted a final design for manual control, leaving NASA with limited visibility into how that system might work. The disparity in progress between the two contractors adds another layer of complexity to an already challenging decision.
Broader contract dynamics also matter. The fixed-price arrangement NASA negotiated with both SpaceX and Blue Origin has been effective at controlling costs and keeping the agency informed of development progress. Yet these very contracts also mean that once design decisions are made, changing them becomes more difficult and more costly. The manual control question will need resolution soon, or it risks becoming frozen into the final design.