Lori Garver, NASA’s former deputy administrator, is famously opposed to reinventing the wheel. NASA, she has written, wasn’t created to do something again. It’s no surprise, then, that she has become a thoughtful critic of the space agency’s approach for returning humans to the moon, which relies on rocketry similar to what NASA used to launch Apollo- and space-shuttle-era astronauts decades ago.

NASA, Garver argues, is at its best when it wields its formidable powers to tackle challenges well beyond the reach of the private sector, whether in exploring other planets or helping to protect our own. Rather than reiterating the Apollo program, she believes the space agency’s most urgent missions should involve tasks such as combating climate change, defending Earth against threatening asteroids and developing transformative technologies for the 21st century. That expansive purview may seem to leave little room for feats of astronautical derring-do, although Garver acknowledges that human space exploration remains a vital part of NASA’s mission. Still, she says, as long as the space agency’s sights are set on off-world horizons, it should at least spend some of its multibillion-dollar annual budget fostering innovations to radically improve how humans can get there.

An Obama administration appointee, Garver was NASA’s deputy administrator from 2009 to 2013. In that time, which she candidly recounts in her memoir Escaping Gravity, she laid the groundwork for the agency’s pivot from a costly reliance on legacy aerospace contractors to collaborations with newer companies such as SpaceX, which offer more affordable and agile launch services. The result? SpaceX has flown cargo to the International Space Station 25 times. Today it’s the only company capable of launching astronauts into orbit from the U.S.—which it has done seven times.

But that might soon change. On August 29 NASA will send its human-rated heavy-lift megarocket, the Space Launch System (SLS), as well as an accompanying Orion crew capsule, to orbit the moon. If all goes according to plan, the mission—Artemis I—will open the door for creating a moon-orbiting space station called Gateway, returning humans to the lunar surface and perhaps eventually sending them to Mars.

Scientific American caught up with Garver before the test flight and chatted about the upcoming milestone, why it matters, what’s at stake, SpaceX’s own Starship megarocket (which NASA now plans to use as a lunar lander for Artemis) and what Garver’s most excited about in space science.

[An edited transcript of the interview follows.]

Artemis I is just around the corner—and for the first time, with a seemingly solid target launch date. What are your thoughts about the program as this launch approaches? How are you feeling about it at this point?

I am conflicted. I want to embrace this point in history that so many people have worked for—to have the capability to again go to the moon. But for me, the opportunity cost in time and in dollars, competing with programs that would be so much more efficient, and the recognition that I do not see a realistic, sustainable path for this program—all that makes it hard.

For the past 30-plus years of my career, when we were talking about going back to the moon, we were always going back to stay. Because we were going to reduce the cost. And it was going to be sustainable. And we were going to have worthy things to do there. I don’t feel like those pieces are in place yet.

Are you optimistic about the Artemis program’s chances for success overall? How many times do you think SLS will fly?

I am asked that a lot. My first answer is: single digits. But I don’t know. It hasn’t flown once successfully. It’s a new rocket, although with old parts, but it’s still complicated. First rockets usually have some tweaks to work out. What’s difficult is that the launch cadence doesn’t really allow for problems, so that might limit its use. But there is a lot of political support for it.

Really, the question is: is Artemis a priority for this nation? Do we feel we must, as we felt during Apollo, get people on the moon in the next few years? And if we do, that current architecture requires a lot of things: SLS, Orion, Starship, eventually Gateway, the spacesuits. There are a lot of moving parts that need to come together, and it’s not clear to me what the long pole in the tent is at this point.

Do you think Artemis is a national priority? Should it be?

I don’t think the public or even political leadership is very aware that we are at least planning to be landing on the moon in three years. More people were aware of the Apollo program, so in that sense, no—I don’t think it’s a priority. And to me, that is fine. The first time you do anything, it’s more visible. But like the first time, we should have a good rationale.

What do you see as the value of returning humans to the moon? Should people care and be excited?

If this program were being done in a way that was different from Apollo—such as really lowering the costs and really advancing technology and being reusable and being sustainable—I think it would be exciting—and not just exciting! These are wonderful, inspirational things to do if they are returning true value the way Apollo did, which was not only to show leadership to the world but to advance an industrial capability [and] advance our national security and our economic vitality by opening new markets and those kinds of things.

For me, one of the things that’s exciting about what’s upcoming is having a broader conversation about what we’re doing in space. I feel what I brought to NASA ended up being of greater value because I had a different background than most people who’d come before. NASA does what it does; it excites the people it excites. They come and work there, and they want to do the same thing.

So I love these moments where we are lifted up—and I think debate is healthy. This is why I’m conflicted about your first question. Because for Artemis, there’s an element of me that is very excited. I love talking to my friends and family who aren’t involved in space. They have no idea what we’re doing, and I wish we were doing it in a way that, when we explain it, people will say, “aha, of course we should be doing that.” We will get more people talking about what we’re doing with this launch, whether it succeeds perfectly or not. It will be a very different conversation, depending on the outcome.

What happens if this initial test flight fails—if there’s a problem with the SLS or if Orion doesn’t make it back safely?

The test flight program is precarious. One of my concerns is that we do not have a test program that is resilient in any way. [Conducting] one test flight and then not being able to launch again for another two years, to me, is not a test. Because if it goes wrong, then what? You can’t put astronauts on the next one, so you gotta test again in two years? A lot can happen in two years, and then of course you have Starship, and that might give the opening to Starship.

Starship is of course meant to be a fully reusable, human-rated heavy-lift rocket, potentially making it far more capable and cost-effective than the SLS. It has yet to fly—though SpaceX may attempt an orbital test later this year. Do you think Starship will work?

That is a very direct question. Typically I say, “If Starship becomes operational, it will be revolutionary,” but that’s not really giving my view. I don’t think I know enough about it other than to say, “SpaceX has delivered what they have said in the past, even if it takes longer.” And I think they will probably deliver on it. I’m not one of those people who just says, “It’s 100 percent going to work.” But I’ll tell you that, [with] NASA investing a couple billion dollars in the lunar lander that requires Starship, it sure seems like the Artemis architecture counts on it being successful.

How expensive is the SLS, and how can it be so expensive? Can you put these numbers into the context of Apollo or the space shuttle?

You can compare dollars, and it all depends on where you start and what you include, but Apollo, [the] full program, was in the $150-billion range. The NASA inspector general has said that Artemis will cost in the [range of] mid-$90 billion by 2025 and that we won’t have landed yet. So each launch being $4 billion when you can only have them every other year compares to a Saturn V, which, I think, in the first five years, launched about 12 times.

More importantly to me is: back then we had no choice! There was no alternative! NASA initially estimated that SLS would cost $8 billion; Congress added another $2 billion. I think the NASA Authorization Act says $11 billion—that’s just for the rocket. Orion was supposed to be $4 billion, then $6 billion, then $8 billion. So together, those would have been $18 billion—and we’ve spent $43 billion.

All that is in comparison, for instance, to SpaceX’s Falcon Heavy that has now launched successfully three times and has about 75 percent of the capacity of the SLS.

This is a very broad and poorly phrased question, but if you can speculate, where do you think NASA would be now if the agency had taken a different path and not moved forward with SLS and Orion? What would be different?

I think we would have ultimately put in a competition, as we have for [NASA’s Commercial Crew and Cargo and Program], if we needed more capability for a larger rocket—and we would have seen the acceleration of those private sector initiatives that we see today. Both Jeff [Bezos] and Elon [Musk] had told us they were already investing in big rockets, but once NASA decided, “We’re just going to do it ourselves,” [Bezos and Musk] had a lot less incentive to get it done quickly.

I think more money could have gone into our Earth sciences programs that people do benefit from, well beyond the NASA employees and contractors. This is the real question: Who is the customer of NASA? And if we can focus more on the customer-advancing missions that have a broader impact, that would be better. And that includes not spending money [on the SLS for] refurbishing engines we already had, building [fuel] tanks we already knew how to build, and so forth. It could have been invested in technologies that would allow us to really lower the cost for launching and doing things in a reusable way and having programs that are going to be more sustainable.

You ask, “Who is the customer?” Aren’t taxpayers the customer?

Taxpayers are the customer. A lot of times we see the customer as a few members of Congress or the very contractors who make money on our programs or the academics who do the research that is funded through the government. It’s a bit of a “self-licking ice cream cone” (not my term).

What are you most excited about in space science these days?

I am, as I have always been, excited about the search for life beyond Earth—so that includes Mars, Mars Sample Return, as well as extrasolar planets that we find first with Kepler and now JWST [the James Webb Space Telescope]. I really bought into the big questions that we can learn through space science. Contact is my favorite of the science fiction movies.

I think that “Are we alone?” question is the most profound one we can be asking, and it’s worth spending a lot of money on.

Yeah, and we don’t even spend a lot [on answering that question]. It’s leveraged pretty well.