SCOTT DETROW, HOST:
I really love space. I love astronomy, deep-space telescopes, missions to the moon, all of it. You might have picked up on this if you have heard any of our regular science segments with the team from NPR's science podcast Short Wave. So we are trying something new today. I'm very excited about it. It is a segment that is all about space news - none of that other science here - with our...
REGINA BARBER, BYLINE: (Laughter).
DETROW: ...Friends from Short Wave, host Regina Barber and NPR science correspondent Geoff Brumfiel. Thanks so much to both of you.
BARBER: Yeah, thank you. I'm excited.
GEOFF BRUMFIEL, BYLINE: Excited to be here.
DETROW: If they let us do this again, can we call it Spacing Out With Scott?
BARBER: Yeah. I like it.
BRUMFIEL: It's your show. You can call it whatever you want.
DETROW: All right (laughter).
BARBER: So today, Scott, we're going to be giving you a quick update on Artemis. That's the mission that's going to send humans rocketing around the moon.
DETROW: OK.
BRUMFIEL: And we'll be talking about data centers in space.
BARBER: And the new movie, "Project Hail Mary."
DETROW: Excellent. So let's start with that. I am both very excited to see it, and I don't know when I'm going to see it 'cause here is my plan. I want to read the book. The book has just been sitting there on the nightstand. I feel like I need to read the book first, then see the movie.
BARBER: No. No.
DETROW: No?
BARBER: No, no.
DETROW: No?
BARBER: Put the book down. Go see the movie. I've actually heard people that that's actually the best way to do it.
DETROW: OK.
BARBER: Let me explain the plot, like, real quick.
DETROW: Please do not spoil anything.
BARBER: I won't. I promise. So in the near future, scientists notice that there are these alien microbes breeding on Venus, then going to eat the sun.
DETROW: Eat the sun?
BARBER: Yeah. Like, dimming it to a point that in 30 years, crops will die, animals will die, humans will die. So world leaders send a mission to a nearby star that's not being eaten, because these microbes are eating all these other stars too, not just our sun. And they want to figure out how to save the sun. So one of the astronauts is scientist-turn-middle school teacher Ryland Grace, and he's played by Ryan Gosling.
(SOUNDBITE OF FILM, "PROJECT HAIL MARY")
RYAN GOSLING: (As Ryland Grace) I put the not an astronaut. I've never done anything. I've never done a space - I can't even moonwalk.
BRUMFIEL: And, Scott, as I'm sure you already know, this movie is based on a book by Andy Weir, the same guy who wrote "The Martian." And he does have a track record of writing really scientifically cool books that turn into Hollywood hits. And, Gina, on that note, I have to ask, is the movie actually any good?
BARBER: Well, first off, Geoff, I mean, it's a blockbuster for a reason. Like, I went in completely blindo (ph). I didn't read the book, Scott. I didn't even watch the trailer. I loved it. I loved the jokes. I loved the puppetry, the visuals. It was all stunning.
DETROW: Wait, wait, wait, wait, wait, wait, wait, wait, wait. Did you just say puppets?
BARBER: Yeah, yeah. You said no spoilers.
DETROW: I know, but I think you...
BARBER: So that's literally all I'm going to say.
DETROW: OK. OK.
BARBER: That's all I'm going to say.
DETROW: But Geoff was nodding at this, and I'm curious. Like, how good is the science? Because...
BARBER: Yeah.
DETROW: ..."The Martian," like, you got the potato growing.
BARBER: Yeah.
DETROW: You got all the science happening.
BARBER: Yeah, a lot was pretty accurate. Like, they did hint to special relativity, where, like, if you're traveling near the speed of light, you're going to experience time dilation. So, like, Ryan Gosling's character aged differently than people on Earth. They did take some creative licenses with, like, the fundamentals of biology, though. So, like, these alien microbes, like, they existed in space. They were, like, on the surface of the sun and they were water-based. I don't know about that. But that didn't seem to bother astrobiologist Michael Wong from Carnegie Science. He loved the movie. He read the book, but did point out that there's one, like, accidental scientific mistake in the movie.
MICHAEL WONG: It's kind of fun to note that the two exoplanets in the movie were exoplanets that we thought existed at the time that "Project Hail Mary" was being written, but in just the past few years between the book's release and (laughter) the movie, we've actually discovered that these planets may not actually exist.
BARBER: Yeah, Scott, so in the grand scheme of things, it's not really a mistake. It's just that, like, science has gotten more advanced since the book was written.
DETROW: All right, Geoff, second topic - launching artificial intelligence into space. Is this a thing?
BRUMFIEL: Yeah. I mean, it's something that some folks are taking seriously, including Elon Musk. He's actually making this the central mission of SpaceX, his space launch company, which just merged actually with his AI company, xAI. Here he is on Saturday unveiling new plans for a massive new microchip factory that would make chips used for data centers in space.
(SOUNDBITE OF ARCHIVED RECORDING)
ELON MUSK: I actually think that the cost of deploying AI in space will drop below the cost of terrestrial AI much sooner than most people expect. I think it may be only two or three years.
BARBER: OK, Geoff, at Short Wave we've done, like, a bunch of stories about, like, data centers and how they use, like, a lot of water. They use a lot of electricity. Like, is this the solution? Like, just sending them up into space?
BRUMFIEL: (Laughter) Ejecting them from the Earth.
BARBER: Yeah.
BRUMFIEL: Yes. Actually, according to people who support this idea, it is, right?
BARBER: OK.
BRUMFIEL: It comes down to really electricity. You know, there is a lot of problems around generating enough electricity for all these data centers on the ground. And if you go up into space, you know, you got the sun and...
BARBER: Yeah, we were just talking about it.
BRUMFIEL: We were just talking about it. Great source of power.
DETROW: Unless it's eaten.
BARBER: Yeah.
BRUMFIEL: It's all free. It's plentiful. And the napkin math looks pretty good, right? You can get as much as you want. But I've got to say, when I started speaking to independent experts, I heard a lot of skepticism.
DETROW: Walk me through some of the specific skepticism, then.
BRUMFIEL: Right. Well, let's start by talking about the biggest sort of computer-y thing we have in space right now, which is the International Space Station.
BARBER: Love it.
BRUMFIEL: It's the largest facility in orbit, and its solar panels generate around 100 kilowatts of power.
BARBER: OK.
BRUMFIEL: Now, that may sound like a lot, but xAI's Colossus computing facility in Memphis, Tennessee, consumes around 150 megawatts. So you can do the math.
BARBER: Wow. I don't want to.
BRUMFIEL: Yeah, you don't want to. Well, that's all right. I actually got MIT aerospace professor Olivier de Weck to do the math for us.
OLIVIER DE WECK: You would have to build just for the power in, photovoltaically, a facility that's 1,000 times the size of the ISS.
DETROW: That seems hard to do.
BRUMFIEL: (Laughter) It's not easy.
BARBER: So big. There's an even bigger problem - right? - on the back end. Like, it's heat, right? Like, space is cold. That's true. But, like, it's a vacuum, so there's no way to transfer, like, heat away from these computer chips, right?
BRUMFIEL: Right, right. And so you need these massive radiators that actually use some sort of fluid to gather the heat from the chips and then conduct it out into space. And those end up being almost as big as the solar arrays.
DETROW: OK, so we started this conversation with Elon Musk saying two or three years away. Then we have an expert saying a thousand times the size of the International Space Station.
BARBER: Yeah.
DETROW: What do we think our real-talk estimate is, like, on those two extreme ends of the spectrum? Is this a thing that, one way or another, you think happens down the line?
BRUMFIEL: I think there's a way to sort of parse this, which is there will be more computing in space for sure, and it'll probably, at least for now, be on smaller satellites, smaller amounts of computing that can be distributed. Whether we end up with giant data centers with kilometer-long solar arrays, I think that's going to be a lot more than two or three years off, but we'll see. You know, Elon loves to give a tight deadline. We'll just have to see where we land.
DETROW: One more thing I wanted to ask about, and this is something that I think is a lot closer than whatever Musk is talking about - I think so at least...
BARBER: Yeah.
DETROW: ...And that is the Artemis II launch. What's going on there?
BARBER: Yeah, Scott. So the U.S. is set to launch humans back to the moon, at least in orbit, right? And that's something that hasn't happened since the early '70s. And this launch window for Artemis II, it opens in about a week from now on April 1.
BRUMFIEL: But there have been some technical problems with this mission's rocket. There was a hydrogen leak which forced NASA to push back this launch. So at this point, Scott, anything could happen. We'll be watching.
DETROW: They were going to launch it in February, then maybe March, now maybe April.
BARBER: Yeah.
BRUMFIEL: April 1. I got a good feeling.
BARBER: April Fools'.
DETROW: If so, that would give us more content for sure...
BARBER: Yeah.
DETROW: ...For Spacing Out With Scott or whatever.
BARBER: Space is pretty vast. We got a lot of content.
DETROW: We got a lot, that's true.
(LAUGHTER)
DETROW: Perhaps infinite. That is Geoff Brumfiel and Regina Barber of NPR's science podcast Short Wave. You can check it out. They talk about science, not just space. Though I like it most when they do talk about space. Thanks to both of you.
BARBER: Thank you.
BRUMFIEL: Thank you.
(SOUNDBITE OF FLUME SONG, "SPACE CADET") Transcript provided by NPR, Copyright NPR.
NPR transcripts are created on a rush deadline by an NPR contractor. This text may not be in its final form and may be updated or revised in the future. Accuracy and availability may vary. The authoritative record of NPR’s programming is the audio record.
