Elon Musk Explain SpaceX’s capability to manufacture, launch, and operate AI satellites at scale

All right, well hello everybody and welcome. Hanging out, I got Elon and Ian Dahl with our Starlink team. Figured we'd check in. It's been, you know, typical SpaceX here. Launched a brand new vehicle. Acquired XAI, now SpaceX AI. Announced a terror-sized chip building project. And so... Yeah, never a dull moment. Typical year. And so let's... kind of wanted to connect some of the dots on how this all feeds into making life multi-planetary. Starting to climb up the Kardashev scale. Maybe show off some cool new AI stat stuff. Let's kind of start galaxy sized and bring people in with the Kardashev scale. What's the big picture? What's the big picture? What is the Kardashev scale? Like, how do you decide what progress a civilization has made? That's the most objective metric that any alien species, say, visiting us would calibrate how much progress we've made as a civilization. And one of the most objective ways to do that is the amount of power that any given civilization has been able to harness. And there was a Russian physicist, by the name of Kardashev, who thought about this. And I think it's a good way to characterize it, which is you can assess how well a civilization is harnessing the power available on the planet. That's type one. And then type two would be how much of the star's power are you harnessing? And then type three would be how much of the galaxy's power are you harnessing? These are very objective and measurable numbers. So right now, we're very low on the Kardashev one scale. Like, if you say, what proportion of our planet's power are we harnessing? It's a very, very tiny number. And basically, we're harnessing almost nothing of our star's power. So the sun is truly an immense thing. It is difficult with words to characterize just how immense the sun is. But this gives you sort of a sense of scale. Yeah, it's a big difficulty jump going from level one to level two. Very big difficulty jump, yes. And level three, and we don't even know how to do level three, really. We'll get there. Yeah, yeah, exactly. AI will figure it out. Yeah. One way to appreciate the size of the sun is to think about how heavy is the sun compared to all the rest of the mass in the solar system? So the sun is about 99.86% of all mass in the solar system. And then of the remaining 0.14%, most of that is Jupiter, one planet. So we're still lightweight. Yes. The entire mass of Earth is in the tiny miscellaneous category. Earth is a tiny dust boat compared to the sun. But how much energy are we talking, like, coming from the sun, especially compared to what we're using here on Earth? Because it feels like... Yeah. The incident solar energy on the cross-section of the Earth is roughly a half billionth of the sun's power output. And the vast majority of that we cannot use because, you know, 70% of Earth is water. Technically, our planet should be called water because there's 70% water. And I think an alien civilization visiting us would be like, why are they calling it Earth when it is mostly water? Were the Greenlands not green of the solar system? Yeah. A bunch of the... Exactly. Even... We're 70% water and 30% that's land. A bunch of it is, you know, Antarctica or, you know, Siberia type of thing. Northern Canada type of thing. Very difficult to... Not places people typically want to live and you're not going to get a lot of solar power at the poles. So the actual usable area of land where you can get solar power is quite small. Anyway, in order to ascend the Kardashev scale, in order to get to any meaningful percentage of the sun's energy harnessed, you have to go to space. If you want to get to, say, a millionth of the power output of the sun, you would have to increase civilizational energy harnessed by much more than a million. So we currently use much less than a trillionth of the power output of the sun. And a trillion is a million times a million. So basically, we're basically practically nowhere on the Kardashev 2 scale. Practically nowhere. So on Kardashev scale, we're all still... We're still failing. We're non-existent. We're non-existent. We're not even... Yeah. We're not registering. Not even Microsoft. Yeah. We're... No. And so to actually... Well, Microsoft would be an epic, epic achievement relative to where we are right now. Something to aspire to. Yeah. Yeah, that's our goal. This is, I think, both simultaneously an incredibly adventurous goal relative to where we are, and yet not particularly adventurous as a percentage of the sun's energy to try to achieve power harnessed being one millionth of what the sun outputs. And so to actually start... Microsoft. To actually start getting there, though, we're not just going to throw solar arrays in space, try to soak up a bunch of the sun. There has to be a need. You want to go up there and do something meaningful. And obviously, until this point in human history, there hasn't really been a need. What has changed to make us think that maybe now's the time to start trying to notch a percentage point or two? I mean, getting to a percent of the sun's energy? Maybe not a percent. Let's go like... We'll move the decimal point back a couple of spots. So you're an extremely kick-ass civilization if you get to 1% of the sun's energy. And I'm like, wow, that civilization is going to be vastly more powerful than us, to say the least. Yeah. So in order to start to make some progress on the Kardashev scale, we need to launch satellites to orbit Earth and capture solar power. And that avoids the need to build massive power plants on Earth and deal with cooling, because cooling is actually much easier in space than it is on Earth. You can just radiate to the vacuum. And so what we're proposing here and what we intend to do is to try to clam the Kardashev scale to be kind of like a respectable civilization. So when the aliens, hopefully there are aliens out there, and they maybe finally decide to talk to us, where we have some respectable amount of the sun's energy being used. That's not totally pathetic, which is the current situation. And so before we start sending data centers, sending all this to space, there are some limiting factors that we got to get there that would traditionally make it so this is almost impossible. Yeah. What does it take to scale? Yeah. So things it takes to scale are you need to have a large mass-to-orbit capability, which is what Starship will give us, that large mass. So you ultimately need to send millions of tons to orbit and beyond. And you need the power associated with that. So if you want to put 100 gigawatts or ultimately a terawatt into space from Earth, you will at some point need a terawatt of solar. And then you're going to need a terawatt of AI chips. So the three things you need are mass-to-orbit, a lot of solar power, and radiators, of course, and a lot of chips. All right. Well, let's start taking down the list. So mass-to-orbit, that's where Starship comes in. Yeah. We just had first flight V3. It was awesome. I know you were there. It was crazy to see that rocket launch. Yeah. And long time coming, what's Starship's purpose of being? What is it going to be doing? Yeah. So Starship is going to revolutionize space, really. It's the first rocket design that is capable of full and rapid reusability. Now, reusability is the fundamental breakthrough that is necessary to make life multi-planetary, as well as to ascend the Kardashev scale. You simply cannot ascend the Kardashev scale unless you have a reusable spacecraft, and you cannot extend life to the moon, to Mars, and the rest of the solar system without a reusable rocket. The cost is simply prohibitive. You can't make enough rockets unless you can refly them. Just like any other mode of transport, you can imagine that if we had to throw away airplanes every time we flew, flying would be far too expensive, and basically no one would be flying airplanes. You'd be doing a whole lot more driving. Rapid reusability. Yes. Every mode of transport is reusable, without which it's simply not viable as a transport system. So cars, planes, boats, horses, bicycles are all obviously reusable. With rockets, it's much harder to make a rocket reusable because Earth has a deep gravity well and a thick atmosphere, and these make it just barely possible to achieve reusability with a rocket. There have been many prior attempts to create a fully reusable rocket, and most of those attempts have been abandoned partway through because they didn't think they could succeed. In order to achieve full reusability, everything's got to be perfect. The engines, the structure, the avionics, the choice of propellants. You've got to go to extreme measures for mass optimization, which is why we have the tower catch the rocket instead of putting on landing legs, which are heavy. The rocket can simply be caught by the tower. And we haven't achieved full reusability yet, but we do expect to achieve that hopefully later this year with Starship. And then you've got to achieve full reusability. You've got to go step beyond that, which is make it rapidly reusable, such that the rocket lands, gets caught by the tower, gets put back on the launch stand, and can be flown again without any refurbishment or laborious inspection, like an aircraft. This is incredibly difficult. This is the first time that there's ever been a rocket where that is possible. That's what makes Starship so profound. It also happens to be the largest flying object ever made, the heaviest flying object ever made, the most powerful moving object of any kind. Starship V3 is more than double the thrust of a Saturn V moon rocket. By version four, we'll be pretty much three times the thrust of a Saturn V moon rocket. And we expect Starship to be flying more than once per hour down the road. One of the fun facts from Flight 12, that was actually the heaviest payload SpaceX has ever flown. And that's still just a fraction of what V3 can do. So, I mean, once we're flying massive amounts really rapidly, I mean, we already fly the majority of payload to space with Falcon. Do people even really understand what massed orbit becomes once Starship is flying? It's many orders of magnitude greater than what is the case today. So even with Falcon 9, Falcon Heavy, SpaceX delivers almost 90% of all Earth mass to orbit. I think somewhere between 85 and 90% right now. And then most of the remaining mass, I think, is launched by China. And then the rest of the world, including the rest of the US, is the remaining, I don't know, five to seven percent. Now with Starship, we'll be aiming to go from somewhere around 2,500 tons a year to orbit to millions of tons per year to orbit. And to do so at a pretty short period of time. So we think probably we can get to a million tons to orbit per year in about three years, thereabouts. Starship is going to take care of the mass to orbit limiting factor. Yes. And then power generation. So first, Ian, maybe you can help. Sure. People probably struggle to visualize a little bit when you say, like, data center in space. Like, we're not going to slap engines on a building and fly it up there. Like, these actually look, like, pretty different. And so kind of walk through how you take something that's in a giant building on the ground and turn it into something that's functional in space. Yeah. I think it's pretty interesting. A lot of people don't actually know what the inside of a data center even looks like, right? Yeah. It's some, like, mythical place where the internet's in the cloud or something. Yeah. Some people envision wires. Some people envision boxes. But, like, effectively it comes down to a set number of chips. And the things that we need to launch into space are actually quite small when we look at it. The more challenging part is figuring out how to get the power for it. And that's where a lot of what we've worked on for existing, like, Starwing technology, the solar arrays, are what we want to utilize that expertise to be able to build a satellite that can actually launch the critical components of the data center into space itself. We like to look at this and say, like, what is the actual engineering problem here? And it's really a combination of delivering power and then taking the waste heat and energy away and sending it into the vacuum of space, as you mentioned. Yeah. Now, the AI satellite is actually much simpler than a Starlink satellite. A Starlink satellite has gigantic phase ray antennas. It's got parabolic antennas. It's got a lot of laser links. It's much more complicated than an AI satellite. An AI satellite is essentially a lot of solar cells, a radiator, and you still need some laser links, but you don't have all of the super complex antennas that you have on a Starlink satellite. So, I mean, given the two, the easier one to design for is the AI satellite. Yeah. It's just a little bit bigger. It's bigger. Just makes stuff bigger, yeah. I was like, so we've got, this is our AI one, if you guys want to walk us through. Yeah, so the first thing that we're really looking at here is, like, first you've got to make something compelling, right? And we thought that the right place to start is around the 150 kilowatt, like, peak power level. But as we look at the workloads with our experience with XAI, we get to actually see that we can also support about 120 kilowatts of average compute. There's a difference. What we're showing here is kind of a draft version of the version one of the SpaceX AI satellite, the AI one, I guess you could call it. And it seems like a reasonable place to start is 150 kilowatts peak power, 120 kilowatts sustained power. And to give you a sense of what does that actually look like in terms of the size of the radiators, size of the solar panels, the assumptions here are 250 watts per square meter for the solar array and about 1,400 watts per square meter for the radiators. So the radiators, these are double-sided radiators that are radiating both sides. They're oriented knife-edge to the sun. And it's 1,400 watts per square meter is a very achievable goal. Over time, we think we can probably do above 250 watts per square meter and above 1,400 watts per square meter for the solar panels and radiators, respectively. But this gives you, this is pretty much what the satellite is going to look like. It's a lot of solar panels, radiator, and then everything else is pretty small, like the virus. And these are like evolutions of things that we have actually already launched in our Starlink constellation to date. That's really, I think, the cool part to me is that we're looking at solar technology that we already are going to use on the V3 Starlink vehicle. So I'm really excited to then just take those and make it bigger. Yeah. Part of what we want to convey here is that there's not some magic that's necessary that doesn't exist for the AI satellites. As Ian said, a lot of this is technology we've already made for the Starlink V3 satellites. So we don't think this is a super hard problem compared to things we already do. There would also be probably something on the order of a terabit of connectivity, of laser link connectivity from the satellite. The 150 kilowatt peak power level roughly matches what, say, an NVIDIA GV300 rack would do. So if you've got a GV300 with 72 GPUs, its peak power, I think, is around 140 kilowatts. But it's almost impossible to be at that peak power. A more reasonable operating envelope would be around 120 kilowatts average power. But it can peak up to 150. So it's basically, think of it as a rack of compute in space. And then you can connect these racks of compute to either each other by the laser links or directly to the Starlink Constellation. So you can close the link with the Starlink Constellation. And then Starlink can then send that data to the ground using the existing KA and KU antennas on the vehicle. It also has laser links to the ground as well. And this would not be at a particularly high latency. We're talking about maybe being around 600 to 800 kilometers above the Earth. And light travels 300 kilometers per millisecond. So that's about three milliseconds away, basically. It's not very far. Won't worry about that too much, though. Sometimes people think there's going to be some high latency. I'm like, no, light moves pretty fast. Light moves pretty fast. It's a tall one. Yeah. I think the cool thing also is the radiators themselves are about the same size as the existing solar arrays for the V3 vehicle, kind of in that realm where we're flying today. Yeah. So, I mean, they got about a 70-meter wingspan. So these are fairly large. And we're talking about building a lot of them and putting them up there. But you like to say, like, space is in the name. Like, there's a lot of space up there. And so even when you're talking thousands or even up to a million satellites, you got plenty of room to move around up there. Yeah. Space is really big. So it's not like space is going to get crowded. Space is enormous. Like, if you zoom in close, the satellite looks big. But if you actually look at it relative to the Earth, the satellites are so tiny you can't even see them. So they're very, very tiny compared to Earth. And I mean, we have about 10,000 starlings in orbit right now. We've got a pretty good idea of how to operate just really large constellations and do it safely now, right? We are the only operator that has any experience of that scale. It's a great thing that, you know, we have this background, so we know how tightly we can pack the satellites and fly them safely. That's a number one goal when we look at the constellation. We're going to be building a lot of satellites. And we're going to be building them here in Bastrop, right? So we've got this, which... Yeah. So we're in that building kind of in the middle, which... Yeah. We're sitting in that building right now. This is my first time here. The building is massive. Like, you come around the corner... And it's big. You see it through the trees, and you're like, oh, wow. But we're about to kind of put this building to shame, aren't we? Yes, we're going to... In fact, we already have the solar manufacturing facility. It's under construction already. And then we will be building out the AISAT production building soon. And yeah, so we expect to have the AISAT production, the solar production and all of that operating at some reasonable volume by the end of next year. So if anybody wants to work on AI satellites, this is kind of going to become the hub of that. We're also... So, I mean, like right behind us, the machines are humming. We're still making all of our user terminals for Starlink here. That's not going anywhere. In fact, we're turning on new production lines for new units, right? Yes. In fact, these are the new Starlink terminals, which we made in much higher volume than the current terminals. And ultimately, we think there's probably going to be a few hundred million Starlink terminals out there. And then the Starlink direct-to-cell constellation will connect directly to people's cell phones and enable high bandwidth communication directly from your phone to space. All right. We're two limiting factors down. We've got mass-to-orbit, we've got putting solar and a few third-ones chips. Yes. So at least in the beginning, we can obviously launch the chips that are already being made. So our current reference design is for NVIDIA Rubin chips, or it could be either GB300 or Rubin chips. And we'll also have a reference design for TPUs. And essentially, you can put any existing chips into orbit. But the current industry seems to be... It seems like it's going to get to maybe around 100 gigawatts a year of AI compute. But that doesn't answer the question of, well, how do you get to a terawatt? That's why you need the TeraFab. Oh, he's looking a step bigger. Yeah. In order to get to the next order of magnitude, you need a gigantic chip factory. And to give you a sense of scale here, we expect that the TeraFab is going to be around 100 million square feet, which is 10 times the size of the Tesla Gigafactory Texas. And what, aside from just, you know, I'm going to need Starship point to point to get from one end to the other, aside from just the size, what's going to make this unique, different from any other chip building operation on the planet? Well, I think over time, there's going to be a lot of technology evolution with the TeraFab. But fundamentally, it's about scale. So even if there were no fundamental technology breakthroughs, and you simply, you could simply scale the existing chipmaking technology with a lot of difficulty to a terawatt of chip output per year. That's, if you look at it just from the Logitech standpoint, that's equivalent, that's like having a billion chips per year with a kilowatt per radical. So it's a billion full radical equivalent chips, each doing a kilowatt. And then you're going to need a lot of memory to go with that. A lot of people today even think orbital data centers were like a decade away. Yeah, I think we want to try to give people a sense of the timeframe. We, at least the timeframe we're aiming for. I mean, you know, people should take this with a grain of salt to some degree because this is just our best guess. So this is not a promise of what we'll do. This is what we are going to try to do and think we probably can do, which is to get to roughly an annualized rate of a gigawatt per year by the end of next year in terms of space AI compute. And then aspirationally scale that by an order of magnitude per year. So in two and a half years hitting an annualized rate of 10 gigawatts a year to space, and three and a half years, maybe 100 gigawatts. And then depending upon what progress there is in chip making in the rest of the world and with the TerraFab going beyond that to scale to a terawatt per year, which is a thousand gigawatts. That's twice the current electricity consumption in the United States. I think there will be an appetite for that, but we'll see. It's a lot of satellites. I don't know what it's going to think about, maybe do a lot of simulations or something. So after we've worked through all the limiting factors, we've kind of topped out what we can do on Earth, what is the next step to, again, try and actually notch maybe some percentage points towards becoming Kardashev level two? Why stop there? Why think small? Because a terawatt actually is very small. I don't want you to think small. Let's not think small. So in order to get to another three orders of magnitude to a thousand X from a terawatt per year, the only way that we can really see that you can achieve that is on the moon with a mass driver. Essentially, where you do local production of photovoltaics and radiators on the moon. Maybe you bring the chips from Earth or you could conceivably make the chips on the moon. But you need most of the mass to be made on the moon, so you don't have to transport it to the moon from Earth. And then because the moon has no atmosphere and only one-sixth Earth's gravity, you can accelerate the AI satellites into deep space without a rocket. So you can basically shoot them into space using an electromagnetic gun, like a rail gun type. I mean, it's basically a linear electric motor as a way to think about it. I think we can show... I mean, if that doesn't get you excited for the future, I don't really know what will. I'm fired up to see a mass driver on the moon. That would be very cool. Yeah, sci-fi future. Yeah, yeah. It would also mean that if we're bringing that amount of mass to the moon, it would mean that anyone who wants to go to the moon will be able to go to the moon. And I think that'd be pretty cool. I'm going to be jumping first in line to get up there. Yeah, I mean, everyone should go to the moon at least once, I think. Yeah, just once. Yeah, you can move there if you want. You can go live on the moon. We'll see. Thanks, guys, for chatting with me for a little bit. Excited to see a whole new kind of satellite, a whole bunch more Starship launches, more chips, more solar, more everything. It's a big future, but I'm excited to see everybody at this company go out and build. All right, sounds good. It's exciting. Thanks, guys.