Aliens, AI, and Saving the Planet - EP 43 Will Marshall

in the next decade we'll see 3 really major things with incredible significance for humanity. One is—

Next decade?

In the next decade. One is AGI, that's— and then second is finding life off Earth, not necessarily intelligent life, but life off Earth, we will discover it. And the third is decoding animal communications. So between chimps, between dolphins, between different animals.. And I think all three are incredible events. And the fact that they may all happen in one decade is— and I call them Copernican events because they're nudging us out from being the center of the universe on everything. So obviously AGI is inventing another intelligence alongside us. Finding life off Earth, even if it's not intelligent, will give us a bigger probability that there's life everywhere and it will have got to intelligent stages. And why do you think we find Well, because telescopes have now got to the stage on the Earth-bound telescopes as well as the JWST are big enough to do atmospheric spectra on planets. So we have thousands, we've discovered thousands of planets mainly by the wobbles of the stars. We look at the wobbles of the stars and go, oh, there must be a planet orbiting around that's tugging on it. Sometimes from them going in front of the star and dipping the light a tiny bit. But we don't know much about those planets. We know the basic size and orbit, but we don't know what their composition is. Now we have telescopes that are just big enough to look at the composition of these. And what we will see is, if you looked at the Earth from the other side of the galaxy with one of these telescopes, or at least these— more like the one— most of these planets are— these star systems that we've detected planets are within thousands of light-years, so the nearby region of our galaxy. You would quickly see that there's about 20% oxygen in our atmosphere. Now, that is highly unusual because oxygen binds with almost anything. That's why we, you know, we get rust and all the rest of it. If we— the only thing that's keeping up that, that oxygen here on the Earth is all the life, all the plants.. And so we could look for that signature or other similar signatures where the best explanation we would have is life is producing it, or more rigorously, if we can eliminate that there's other chemical ways or geological ways, non-life ways of producing those things. So, so you, you think we've just had enough advances with the, the size of the telescopes to be able to block out the sun and, and just focus on the planet and resolve that and do spectra on it? We've just got to telescopes that are big enough to take spectra of nearby by planets. We've got all the targets lined up, we've got thousands of planets, we know where they are. We now need to look at them with these new next generation of telescopes. And I think, uh, we— it's just the chances of us— I, I think we'll get to the point of having a signature that we say the best explanation by far is life within the next 10 years. It won't be certain, we won't have visited it, um, we wouldn't have had a little green man wave in the telescope if If we can talk to animals, I feel like it's going to go really bad for us.

I mean, fascinating, of course, and would 100% want— but those first discussions, I don't think humans fare very well in what the animals have to say back.

They might be pretty pissed off. Yeah, since we just wiped out most of them. But I think it would be fascinating, right? Like, what are they really talking about? What are they— And there's incredible— I mean, whales have been talking, if they've been talking, they've been talking for 70 million years. Wow.

They know stuff.

They know stuff. Like, can you tell us anything about that? Just imagine the culture that we could learn about.

You think we're going to know what they say to each other?

To each other.

But we won't be—

Well, no, then we can talk to them for sure. I mean, like, it's an amazing precipice that we could And then of course, I hope that it will expand our empathy circles to include them in the things we value. Right now, we don't really value life.

Pretty, it's gonna be a lot harder to eat them when they're—

I hope so. I really hope that's one of the consequences.

Yeah.

Yeah.

So, and you think that happens because AI helps like pattern match everything that's going on?

Yeah, exactly. You're gonna feed it all of this acoustic data and maybe even body language data or what have you for chimps and it will start deciphering. And it may not be that sophisticated. Maybe the dog just says, "Hey, I'm hungry," or "I'm pleased to see you again," you know. But I think we will discover some subtleties about their social behavior about— and we'll certainly empathize more with their plight, even if it's a very simple— it might be like listening to toddlers that just start understanding language, right? So it's very simple, but it's interesting to understand their worldviews and what they see as interesting, what they question. And, uh, yeah, I think those three moments are all decentering humans from the middle, just like Copernicus, uh, decentered—

oh, I see—

and then Earth.

But if we— and if we saw life that far away, it's not like we would be able to— like, it would be amazing to know but we're not getting there anytime soon to have—

going from one to more than one life, the genesis of life in the universe will mean we would know a lot more about the statistical probability of life and how much it's arisen, right? And we've got thousands of planets. If, you know, a handful of them have life, we would then know that the universe is teeming with life. Um, On the other hand, if we don't find any, or none with even strong suspects, because it will never be black and white, it will be an evolution, as I said, that we'll probably find out this planet has got this chemical signature, and the best thing— we can't think of a volcanic reason that we would have that. The only obvious answer is some sort of life. We don't know for sure, but that's the best explanation we have. We'll get to that point in this, in this decade. And knowing that um, that is cosmologically significant because again, we've only had one example so far. And if we discovered it on Mars or nearby planets, then there's still the question of whether it was panspermia, whether that is, they— there was one genesis of life and it's spawned onto the other one, because there's a lot of material moving between our planets in the solar system. But if we discovered it on another solar system, the chances of that are really diminished. So it's almost certainly an independent genesis of life, which will again mean probably, since we've only sampled a tiny, tiny, tiny, tiny fraction, um, I can't remember exact numbers, but roughly I remember, uh, thinking there's about a million million million or 10,000 million million million Earth-like planets if you just rounded it down to things that are about the size of the Earth with the chemical composition of the Earth in the universe. So We've got just God knows how many planets of all different shapes and sizes, but we just don't know the fraction of them that have life. We know that there's gazillions that have the potential to host life if life got there, but we don't know how many on which life independently arose. And we don't know that probability. This main missing question in the Drake equation of how much life is in the universe is the 0 to 1 bit on, on if we— if the conditions are right. Uh, uh, so we know there's lots of places where the conditions are right, but I don't know how often it's arisen.

I mean, if all three of those things happen in a decade, yeah, I don't even know what to begin to tell my children to like think about, to study, to navigate. You know, I just feel like we're heading to such an upside-down—

It's gonna— I mean—

Fast-moving.

Yeah, I mean, my gran said it was amazing because she saw from horse cart to man on the moon. But this is much more dramatic, a change, uh, in terms of the intelligence explosion and uncovering, and, and in terms of the— our position and place in the cosmos, more profound than any of those things. Yeah.

Well, given the breadth of things I cover, there's a non-zero chance I could break at least one of those stories, I think.

Yeah, for sure. Yeah, try it. Go for it. Yeah, I, I think it's just— that is tantalizing to me what's happening in the next 10 years in those worlds.

Like, that excites you? It doesn't scare the living shit out of you?

I think it's both, uh, it's, it's— but mainly it's excitement, especially on the aliens. I mean, I can't wait to meet the aliens. No, but I mean, as I said, it's most likely non-intelligent, but, but still, knowing that there's life off Earth will be such a huge thing about life being teeming in the universe.

Will Marshall, what a treat.

It's great to be here.

Unlike most of my guests, I have written a book about it.

Yeah, you know a thing or two.

Made a movie about it.

Yeah, what else do you need to know? I don't know why I'm here.

Well, this is so other people could get to know you. I usually am plugging someone else's book, but I get the rare treat to plug my own.

There we go.

That's my plug. When on Sale. Yeah, do you do?

I'm just kidding. No, of course.

Do you have one? Yeah, I do. Okay, I should make that happen.

Okay, so—

I'm trying to embarrass you. To give people who might not have read When the Heavens Went on Sale or seen Wild Wild Space— shame on them— on HBO, I'll give some context. Will Marshall, a man who's hard to describe in some ways, a physicist, a quite accomplished physicist, obviously the co-founder and CEO of Planet Labs. For people who do not know what that company does, they really pioneered, um, a new way to image the Earth from space using hundreds of satellites that, that take photos of every spot on the Earth's landmass many times a day. It's been a revolutionary technology, not just in terms of getting new imagery about Earth, but also in how satellites were manufactured, turning that into a mass production, um, feat, which now we see SpaceX, other companies doing all over the place. And so, and worked at NASA, did a bunch of cool missions, one of which not only like helped confirm that we had water at the poles of the moon, but, but found more water.

Lots of water.

Yeah. Lots of water.

5 or 6% by mass water on the south pole of the moon.

Yeah.

But also like hydrocarbons. So there was methane detected, CO2, some nitrates, really interesting stuff in volatile form that could be useful for a base there one day.

It was a really cool mission. You were part of a— which we document in the book and the movie— this revolutionary group of youngsters at NASA turned NASA Ames into quite an exciting place.

Yeah, under the stewardship of Pete Worden, who deserves all the credit that you give him in that book.

He's a great, great man. And then also One last bit and then I'll shut up. The thing that I've been interested in for a long time, lived in communal houses, kind of a communal house pioneer in Silicon Valley, which we will get into. So yeah, those were the big bits. Are you offended if I left anything out?

Not at all.

Not at all. There's tons of directions I want to go. First off, just as far as the book and the movie go, I always try to give people a chance to settle, you know, since I get the pen and I get to make the movie. Anything you would like to publicly, you know, flog me about?

No, not at all. Firstly, I thought it was incredibly delightful that you chose this topic to dive in a little bit more. A lot of people don't know about this revolution that's happening in space. I thought the way you did it was a relatively good character portrayal of all the crazy kids there, you know, like Chris Camp. It's not necessarily an exaggeration what you say, you know, and Peter Beck. And, you know, I think you get the— you know, I've challenged you before on the title, which I thought didn't really— and your editors pushed you on that. And, uh, um, but it's not about the commercialization. That's not the focus. The focus is so much more deeper, missionary, um, and, um, it really is about how do we help life on Earth, how do we explore the stars. It's, it's so much bigger. To put it down into monetary terms is to belittle the ambition in a way of the crew and the cast of characters. I know you had editor pressure But I want to tell you one fun anecdote that after the Wild Wild Space, I was at a very high-level conference, the Munich Security Conference, a lot of heads of state and generals, and it's the big security forum of the year. And a general literally chased me down because he recognized me as I passed him by in the corridor. And it was basically the Deputy Supreme Allied Commander of NATO. And he had seen the Wild Wild Space documentary, got so excited that it, um, um, he, he wanted to have a meeting and talk about how our data could be very helpful to monitor Russia and other things that he was trying to do and to demonstrate how NATO could move faster on, on new tech. And so we ended up forging a partnership that actually got announced, uh, earlier this summer. It was a great, great partnership that we're still at the early phases of, but it monitoring large areas with AI and our dataset, but it came directly. I cannot explain any other thing that instigated that partnership.

I'm sure he would've found you anyway. All right. What do we do at Core Memory? We cover innovative, fast-moving, forward-thinking companies, which is why Core Memory is sponsored by Brex because Brex is the intelligent finance platform for many of these companies. 30,000 companies from startups to the world's largest corporations rely on Brex's technology for their finances. They've got smart corporate cards, high-yield business banking, and expense automation tools that are fantastic. I hate doing my expenses, and Brex's AI software runs right through those expenses, figure out where we're spending money. And take care of so much stuff for you so you don't have to waste your time on it yourself. Go to brex.com/corememory to learn more and just, you know, get with the program. Let's get going. Let's get out of this archaic finance software and move toward the future. CoreMemory and Brex. The books, it seems to have like found an audience among the military. The Space Force made it like book of the year or something. They wanted me to come speak in Colorado. And, um, I mean, I guess it makes makes sense, but in some ways. But yeah, it's always funny to see where these things end up. Have you had other people recognize you on like a United flight or anything like that?

So many people started saying, I saw the film because United put it on there. Yeah, lots of people. And yes, occasionally, yeah, I've had a few. Luckily it's not too much, so, but like a few times I've been hiking out and somebody's 'Are you that guy from that space documentary?' I'm like, 'Yep, guilty as charged.' I know you're upset about the crass capitalism of the, uh, title of the book, but—

and we will get into that, I think, with some other questions. But, um, you— and you were a— what I was curious about is you were a somewhat reluctant participant in the documentary. It took, it took a while to, uh, warm you up. I think my sense was always you were a bit, um, unsure of the outcome on that. Do you think that's fair?

And I think it's fair. I think that my logical brain is always, must do the most important things to get our mission accomplished. And I didn't put media plans as center to that. So that's one aspect of it. Secondly, I'm like, I don't want people prying into my personal life so much. It's just I'm not trying to, you know, get attention like that. But I thought you guys did an amazing job in the documentary of portraying what was going on. And I was so pleased and proud of the way which it showed how these entrepreneurs are doing amazing things. I thought so. I think you did a great job. All right.

You're being nicer to me than I expected. Okay. You have to tell me if this one story is true or not. True or false? You do live and have lived in a community communal house for many, many years. We asked many times to film at the communal house. You told me, I think at one point, that the house had voted no, they didn't want the camera. Then, however, I was at a party with somebody who lives in your house, and they said, "Yeah, it's too bad you guys didn't film." Really?

I didn't say all, it was universal.

Well, they said, I said, "Well, Will didn't, he didn't want that." And they said, "Well, we took it." said, asked us to have a vote one night, and we all voted yes to, uh, to film. And now I need to know if that is true.

Honestly, I can't remember doing that, any such a vote. Um, but, um, but definitely there were varied opinions. But no, it was more personal decision for me. Let me just own it. Like, it's more personal decision for me.

I want to start with the communal house bit just for a second, which is maybe not the most natural place to start, but I think it's interesting. So you guys, I, I think it is okay to say that you used to live in the Rainbow Mansion. Mm-hmm. Which is, was this large mansion in, in Cupertino. Still exists today. There's still people living there. I think this is, these kinds of houses, I guess, go back decades in some ways to the, the, especially in the Northern California spirit. Um, but I always felt like you guys put a modern twist and flavor. A lot of it had grown up out of the counterculture movement and, I don't know, some culty kind of things and stuff like that. But in this period when you were working at NASA Ames and were with a bunch of these smart, energetic 20-year-olds off doing interesting things, you, you had a group living in this space. I always felt like you took this to the And I don't— anyone who knows me knows I'm not saying this in a derogatory way, like the nerd version of this house, you know what I mean? And it shifted a little bit. And then in the subsequent years, these communal houses of tech people are all over Silicon Valley. I find a lot of times they're, they're almost themed. It's like, this is the effective altruism house. This house is doing hardware, whatever. And so for people who have never lived in a communal house, who have no idea what this culture is like. You've done this now for what, like 25 years or so?

20 years. Yeah.

Yeah. I mean, what is it like? Why do you do it? It's an important part of your life. I mean, you guys like believe in this deeply.

So I like to spin the question around of why wouldn't you? So it goes a bit further back than the '60s hippies counterculture. Of course, it's how humans have lived for all of our evolution. And the more, uh, the more obvious question to me is, why wouldn't you live in a community? And in this case, you know, we— that— so I can give you a practical answer, which is, you know, we feed on ideas off each other. You learn what the latest stuff in AI and robotics or nanotechnology or what have you by having a community of people that. And then you see intersections and then you spur each other on and you support each other in doing those great projects, trying to help the world. We did a— it was, I think, the technology is one thematic focus, and another thematic focus for our community houses has been technology and society and the intersection of those things. And how do we bring about technology into society in a way that maximizes the benefits, minimizes the harms? And we thought a lot about those questions and I would say, uh, so I don't know how much our community houses have directly led to other community houses in the Bay Area. There certainly feels like there's a big resurgence of them. I'm not sure if that's because of us, you know, I can't attribute, um, that. But, um, what I could say is that, uh, that, that, that the ideas that have come out of these houses have been absolutely widely shared. I mean, I always think that We are talking about the things that start to become popular to talk about, about some of the challenges of technology, for example. We're talking about them 5 years earlier, at least several years earlier.

Because you have these communal dinners, you have these salons.

Yeah, philosophy salons where we talk about, okay, what's going to be the application of this cryptocurrency when combined with this humanitarian situation over there, combined with it. And we talk about those, we scenario that, and we have interesting thought leaders come through and they know when they come to the Bay Area to stay at our place, because they've stayed there before and they know that that's where they'll get the latest touch on what's going on in Silicon Valley. And that community has led to ideas that have, you know, numerous TED Talks on numerous topics have happened a few years later on that topic after we've baked it a little bit, if you like. So really, I think like, so, so yeah, but, and that more philosophical question is like, why, you know, why would you not live in a community? Um, I think the, the, the recent and local phenomena of living not in a community is a peculiarity, uh, that will get reversed when people realize it's very natural to live with other people, and especially a chosen family. We call it a chosen family because these are people that effectively are brothers and sisters because you live with them for and do cool stuff together with them. And, um, but, um, to do that, um, it's so much more fulfilling than living on your own if you can get basic things right. And people immediately think, oh, it must be a mess, like student dorms. It doesn't have to be a mess.

Just, you know, seems like you would be at odds with someone and it would be very awkward to get them out of the house. And I know you've met people. Yeah. But it still feels like that can, that could go wrong often.

It can, not often. We've had more than 100 in our community house and once or twice we've had issues where we really actually had to ask somebody to leave.

Yeah. And you, even as you've gotten older, you look, not to be crass again, but you've made quite a bit of money, I assume, from Planet. Your tastes, your desires, they haven't changed over the course of these 20-plus years?

No, I don't really care about shit like that. And ironically, once I realized I was— because I don't look at those detailed finances all the time, and one of— I suddenly realized that somehow that I was in the smallest room in the house paying the most because we had set up this egalitarian system of paying, which was dependent on your salary, not your— anyway, your ability to contribute. And, uh, but I liked having more people, so I had chosen this really small room that no one else wanted.

But are you still in that room?

No, no, no, things moved on.

So what is your room? Is it— does your room look like a luxurious, uh, bedroom, or is it closer to a dorm or something like that?

Well, I've had various rooms over the years, and Uh, yeah, so, but yeah, I don't know, there's not much to say about that.

I mean, do you, at one time, didn't you sleep on a bit of plywood over a bathtub?

Yes. Next question. Yeah, no, I mean, how long did you do that? Like a year or something like that. It wasn't that long. Uh, yes. Um, I was thinking of making a contraption to let me fall into the bath in the morning as an automatic alarm clock.

Did you have to take the plywood off so people could use every day?

So people, no, no, no, it wasn't an active bathroom. I converted the bathroom into a bedroom. But yeah.

Okay, I mean, I think you do have a point, which is it's bizarre to me in our neighborhood. Like, I know almost— I know one neighbor, but I don't know anywhere near like all our people. People just don't talk to each other.

We have lost community. Yeah, when we threw out churches, uh, which, you know, I have some sympathy for in certain senses, um, the fundamental dogmatic belief systems and so on, we threw out the baby with the bathwater. The baby being the community sense and the time together to think together. And, you know, someone giving a sermon about what's right and wrong and turning the other cheek and all this basic stuff, and we don't do that anymore. We don't have a community, we don't rely on each other, and we don't— and I think that's part of the challenge we face in our society. We've got a lot of divisive politics and other things because people can't talk to each other.

But you trace this to, to religion being less of a—

well, certainly in Europe, which is where I'm from, um, average attendance in churches just plummeted in the last sort of century. I don't know exact numbers, but the vast majority of people don't go to church, but because they're not religious, they didn't subscribe to any particular religion. But that, as a consequence, lost the community. Okay.

And what, why do you think, I know you guys have, you've even started some communal houses in other parts of the world, but why is this primarily a Bay Area phenomenon? And if so, why?

Well, again, I, I'd like to turn the question on its head. I think it's a fairly Western and local in time phenomenon that people don't live in community cuz in much of the world. Often grandkids and uncles and aunts and cousins are all mixed together in bigger, much bigger family units, at least more like we have throughout history. And it's only really in the West in the last 100, 200 years that we have separated out to this nuclear family unit concept and try to live on our own, which made it much harder to raise kids because it takes a village and all these other challenges that I think that stresses on society.

Okay. In the context of Planet Labs, I mean, the Rainbow Mansion played a key role, right? So some of these discussions that you're talking about took place, this idea of— you've always, you've always been an environmentalist, I think it's fair to say. And this idea that one of the original ideas was, okay, we're going to photograph the Earth in ways that haven't been done before to show people, to keep track of what is happening to the planet, to see when things are being deforested, to see how rivers change, to monitor. And, and that did feel like a direct result of the conversations you guys were having, for sure.

Yeah. And many other startups, uh, came out of these community houses as well. Um, but yeah, for sure. Um, uh, I could also trace some of our lunar missions that you mentioned earlier to, to some of the conversations that we had in, in Rainbow Mansion. Um, But yeah, my personal motivation is strongly related to how we can help track all of nature to help us to take care of it. The— if you think cosmically, like if you think the big picture, life is either extraordinarily rare or singular. Uh, I think it's much more likely the former than the latter, but we only know of one location for life, so And I did a quick calculation the other day just to give you a sense. Life matter in our galaxy, and it doesn't seem like there's— let me get the calculation right. So we think of gold as pretty rare on the Earth, and it's about 1 in a few billion atoms per, you know, a billionth of the mass of the Earth. Life in our galaxy, if it is on 1 in 10 star systems, is a billion times rarer than gold is on Earth. Life is just— even if it says the sort of maximal end of, uh, of how much life there could be in the universe based on our current observations, or at least in our galaxy, let's, let's— if there's not some crazy hidden structures of massive amounts of life in our galaxy. It doesn't seem like there is. It's a billion times rarer than gold on the Earth. Like, there's basically— the vast majority of mass of the universe is gas and dust and rocks and things that are not alive. The preciousness of this, just from a diversity standpoint, is obscene, right? So we have this incredible diversity of life and, uh, it's almost our cosmic duty to take care of it. It is cosmologically significant. It's not just locally significant, you know, it's cosmologically significant because it's either so rare that it's just extraordinarily, extraordinarily rare stuff in the universe, or it's the only one. Either way, the duty of humans is above anything else to help take care of life on the Earth. The incredible complexity, the ecologies that we find are so beautiful. We are so lucky. And yeah, so it's also a cosmic tragedy that we're— and this is uncomfortable to talk about because people don't like talking about non-fun stuff, but we have wiped out 70% of life by numbers, by population decline. It's 80% for wild mammals. So there's 80% less deer and moose and And there's 70% less fish in freshwater rivers and lakes. There's 75% less insects than there were 40 years ago. There's half the corals have gone, half the forests have gone. We're doing what on this most unique and rare thing? Imagine taking your one piece of jewelry and smashing it with a hammer and throwing away half of it. I mean, come on. Like it's, and Um, so yeah, I think it's our most important priority and we are not doing a very good job right now. And he, so yeah, part of the point of Planet is to give ourselves the tools to help us to take care of that.

Okay. Since day one, the Core Memory Podcast has been supported by the fine people at E1 Ventures. They are a young and ambitious VC firm in Silicon Valley investing in young and ambitious companies and people. Thank you so much to E1 Ventures for all your support. And in that context, I run the risk of sending this podcast off in a drastically tangential direction. But some of— a lot of these issues that you brought up, you and I have talked about many times. I haven't really caught up with you since the whole AI boom, this part of the AI boom took off. When you— and I feel like sometimes this question is so, um, can lead to— everybody has their P-Doom and all this, but I haven't talked to you since all this stuff happened. I mean, are you— where are you? You were already concerned about where the planet and humanity was going. Are you— has AI made you feel better or worse over the last like 2 or 3 years?

Well, it's a complicated question, but AI, I'm very excited about, and I want to get into some of the things we're doing. Yeah, I think it's And just as a person, I mean, it's incredible what is now possible, right? Almost everything we say, oh, only humans can do this, we can now do it with AI. I mean, amazing. And, uh, but just in terms of its interface with climate and the environment, um, there's this thing called the Jevons paradox, which, you know, basically, um, but everyone knows that AI is eating lots of energy directly, right? It's driving lots of energy consumption, and that's going to massively increase. But I think the bigger impact for the planet is its indirect implications that it's making everything more efficient, whether that's agriculture, oil and gas, mining, all these big industries. What is making those industries more efficient? Well, that mining operation that wasn't going to be possible before, that oil and gas exploration that wasn't going to be possible before because it was just slightly too expensive, now it's just slightly possible to do. Do. And now, so in general, when the Jevons paradox is sort of version of elasticity of demand, humans, when they invent new technology, things become more efficient. You could in principle use that efficiency to do the same thing with less. I can make this can with slightly less aluminium. So I will use the same number of cans, but with less material. Great. But in practice, what humans do is then go and use that efficiency to do even more. So then we end up extracting even more aluminium than before because we just have twice as many cans. So the AI is that on steroids, right? It's making all sectors of the world more efficient. And that means we're going to speed up the blowing up of the planetary boundaries. Our wasting away of biodiversity is being driven by material consumption, agriculture, and these things. So by making them more efficient, we're gonna drive that in faster. So whilst AI can help us in key ways, climate modeling, helping us solve fusion, helping us make more efficient use of data centers, grids, the more impactful, unfortunately, I think, This trend, the default, um, way in which AI is going to, uh, have an impact in our capitalist-driven society is definitely going to be accelerating the exploitation of the planet unless the AI leaders themselves decide, or we as a society decide, to deploy it preferentially on that climate modeling and fusion and less preferentially on resource extraction industries. But I don't see any signs of that right now, unfortunately. You know, so, so meaning unless it's like one of those technologies, all technology is how you use it, man. But by default, it will accelerate us blowing up the planet. That is pretty clear to me. So obviously we're trying to help the reverse by merging AI and our datasets to help us take care of the planet. But if you, if I zoom out and think about the macro, how is AI going to play with with our entire economy, I think that's the way it's going to play out.

All right, well, that was bleak. I, I was curious where you— I, I actually thought you were gonna go like hardcore optimist on that one.

Uh, not on how it interfaces with, uh, the planet.

No. Yeah. Okay. All right, well, let's get into your technology. I, and I'm gonna do some of the context and, and and then let you take it away. For people who don't know, we've had imaging satellites for decades. Actually, I think the first ones were built over here in Sunnyvale in the '50s and '60s, the Corona Project. They tended to be built by governments. They were giant machines that were expensive and rare. The people taking pictures of the Earth with the same thing. They had— it was a small group that had this, this access to what was going on. They could only look at a handful of spots at any given time. And then you guys had this big idea to blast it open. Yeah, instead of making a few of these very expensive satellites— I always think of it a little bit like when we moved from mainframe and big Unix machines to these clusters of Linux machines. What, what kind of gave birth to Google was you guys split up the job among a lot of small satellites and then surrounded the Earth with hundreds of them. And so you were able to do a couple of things. One was not just have all your eggs in one basket on this one $1 billion satellite, but to send up lots of cheap ones and then to clearly surround the Earth with this. This— they just— they're in this group and they're just going around and around. So you're getting not just pictures of a few places, but, but of everywhere. And you're getting this historical record keeping. This was a huge idea. It is a very democratizing idea because mostly it was like the United States and Russia and China and a handful of governments that could usually afford these kinds of things. And now you had a service that people could access on the internet and tap into and get these images. So it was amazing. Like, and I still don't think most people, enough people understand that this shift has happened over the last decade and we're in this new world. Tell us, you know, I got burned out after I did the book and the movie, so I'm not up to date on the latest and greatest of like all of the planets. I can't believe it.

I thought you were jiving me for not listening to your podcast episodes.

Planet stats. Tell me, give the audience a feel for how many satellites you have, how much of the Earth you're capturing every day, what types of things you can see with your different satellites.

Sure thing. Yeah. I think of it as a global sensing revolution., and we're in early days of it, both in the sense that it's obviously still getting better and better, and even more so in the sense of the exploitation of it, which is, you know, I would say less than 1% into a big future, and AI is accelerating that, and we'll come back to that. But on the sensing side itself, yeah, we have the majority of our satellites, we have about 180 of the Doves or Super Doves that image the entire landmass of the Earth once per day. At about 10:30, 11:00 AM local time, no matter where you are, and some coastal waters and sea areas. And they do that in 8 spectral bands at 3-meter resolution, all in optical and near-infrared. Then we have a separate set of satellites, about 15 satellites that we acquired from Google, which is the Terra Bella acquisition, 50-centimeter resolution, so higher resolution, but tasked to note specific locations of interest. And then one thing that's totally new since we last caught up, we had— we've just launched a first satellite that is a hyperspectral satellite. So this has 420 colour bands, just like our eye has 3 colours. We have 420 colours spanning a wide spectrum. So starting at 400 nanometres out to 2.5 microns. Um, which they say if you want to, um, get a quick understanding of something, take a picture. If you want to scientifically measure it, take a spectra. And it can basically tell us what the kind of chemicals things are made of on the ground. It could tell us the type of species of tree or coral reef or, um, grass or what have you on the ground, the shade of paint. So well for defense and intelligence applications that they can say this tank was made at that factory because their shade of green or whatever it is, camouflage is slightly different from the other factory. And so it's like a fingerprint of objects on the Earth. And one of the amazing things that can do, by the way, is detect gas emissions. So, turns out methane gas, for example, absorbs a particular set of light in a very narrow spectra. And so by, by looking for that and the ratio between that and other spectral bands nearby, we can actually tell the volume of methane in the column between the satellite and the ground. And so you can measure the amount of methane gas leaking from a facility like a gas leak or cows farting or, or from, from a landfill site or whatever. So that's a new thing, a new tool in our arsenal. And we did that with NASA JPL and a consortium of players. It's the most exquisite, it's the most sensitive hyperspectral instrument in the galaxy, according to people from JPL. And I don't know how they determine that, but it's cool.

And you're— these images, it is cool. And so you can obviously track all kinds of things. I mean, you could watch, you can watch construction projects taking place. You can watch, check to see— I know there's in the Amazon, used to have to have people on the ground to try to discover when some area was being illegally cut down. It was not an efficient process. It was a dangerous process to go confront these people, report on them. Now you have a satellite equipped with AI software that just like instantly detects when something like this happens. You've got the methane bits and pieces, obviously. Well, and then agriculture is a huge market, knowing if crops are healthy, when to harvest them. You guys look at like the chlorophyll, I think, in the plants and can determine this from space. It's all of it is extraordinary. The, over the last, I always, there's a frustration with this to me to some degree. I mean, you guys are a public company now. You're making tens of millions of dollars per quarter. The people are using this technology. They know about it. You have all these huge deals with the governments. When I see the difference in the size of the business between imaging and communications with something like Starlink, And going back to something I said before, obviously there are huge groups of people who know about this technology. I guess I'm sometimes I'm frustrated that it feels like you said 1% that we've only scratched the surface with some of this stuff because it seems like such valuable information. I think I talk about it in the book like a Google search engine of the planet and this like real-time accounting system of what's happening on Earth. Are you surprised that Is this the pace at which you thought people would know about and use this technology, or has it gone slower? What you— yeah, I'm always—

I definitely would hope that it would go faster than it has. And in the future, but we're, you know, the difference between comms, you're swapping someone from one comms system to another slightly. Here we're mainly market making. So most farmers, even though in principle we could, uh, help them, in practice they don't know about it. They don't know to use satellite technology. They think that's far out. They, or if they've even thought about it. Um, so when you're market making, it takes a little longer. There's a great article about data companies, and I think of Planet as a data company with some cool backend called the satellite, uh, just like Google is a, you know, not a server company. People think of it as a search company or whatever. But that, you know, as that little article talks about how, what the differences between a data company and a software company and a software company and a hardware company. And data companies tend to take longer to build up because they have to build out the market, but they tend to be a winner-take-most sort of situation. Can have very high margins and very high lock-in once you're there. You used to work at Bloomberg. Bloomberg, I think, is one of the— we got the—

they had the ultimate lock-in.

They did. It's a data company.

Yeah.

Not the bit you were working for, but the terminal, Bloomberg Terminal, the part that made money, the part that made money that paid for your salary was supplying data to financial analysts to help them make smarter decisions on their hedge funds or their insurance or whatever. Uh, we're doing the same, providing data feeds. That took them a long time to build up, but now it's a cash machine, right? And it's extreme. In fact, the banks tried to get together to unseat them and had a real hard time, even if they undercut the price, because the lock-in of having a data set that's already deep into your workflows. So look, um, what I'd say is the applications are huge. We estimate the market to be 50 to 100 $100 billion a year just for our Earth imaging data and analytics sitting on top of it. And do I wish we got further faster? Yes. I mean, I'm Mr. Impatient, literally, at our company and always going around. But I think that we're now, because of AI, at the takeoff of it. And I say that with some trepidation because I can't predict the future, but I really believe that we're— that is bridging a gap. There's an awareness gap, and then there's a solution gap. That farmer, imagine you're a farmer and you're trying to improve your crop yield. If we give you some pictures, that's not very useful. But imagine if we can, as we now do, automatically extract out your field boundaries of your different fields, automatically say which crop type, is it corn or soy, and automatically figure out which area of the farmer's field, your field, is doing less well, needs water, needs fertilizer, needs help in some way and helps you do precision agriculture that improves your crop yield by 20 or 40%, reduces your fertilizer use, which is good for you. It reduces your cost and it stops the flow of, of those pollutants into the nearby river. Good for the planet, good for everyone. And we've aligned our business model so that if that farmer wins, we win, um, and planet wins. And so We've got the engine there. That farmer needs that analytics to bridge that gap. But once we have that, the scaling potential is huge.

Is there something that changed? Obviously Transformers came out in 2018, I think, and then kicked off this wave of immense progress in the AI field. But I, you know, I remember there were companies like Orbital Insight that were trying to do similar things to what you're saying. I don't want to speak out of turn. I think they might not be around anymore. They got bought. So, so, you know, and they had trouble kind of bearing out this business on them.

Absolutely.

But we're on it very early. Has something changed in the last couple of years to make that type of analysis more, you know, either productive or useful, I guess?

Yes. I mean, incredibly so. So, We've been working with AI for years, but using convolutional neural nets, CNNs, that enable us to bespoke train a model. We do building detection, road detection, airplane detection, ship detection. We provide those feeds. You can circle ports, get reports of the number of ships in these ports. I can give you some concrete examples, but what the LLMs are doing because of that Transformer development, um, or as I like to think of it, the foundation model work, because they've now broadened out from just language to be multimodal, um, so they can take in audio or visual imagery or video. And as they went multimodal, the relevance to satellite imagery is just huge. It means you can automatically search for things without having to have trained any model. You can show a satellite image to Claude or Gemini or ChatGPT, and it will have a fairly good description of it out of the gate. And then you can show it a picture from a week earlier or a year earlier and say, what's changed? And it'll have a fairly good description of that without having to have any expertise. I did— our team was looking at this, and we Like, can it just estimate land use change over time? Plot some pie graphs of the changes. Um, yep, no problem. Can it count new roads or buildings? No problem. And so in 2018, I'd done this TED Talk, Queryable Earth, where I said, uh, Google's mission was to index the internet and make it searchable. Planet's mission is to index the Earth and make it searchable. The confluence of— so that's, that was, yeah, 7 years ago. Now LLMs basically make that really tractable, this queryable. And just imagine all these LLMs that you play with online, which are really cool, are basically restricted to the information on the internet. So you can answer your questions about the text on the internet, basically. Um, imagine that we suddenly pull in the physical Earth and now it can answer questions about the physical Earth, about all the trees, about the coral reefs, about the roads, about the vehicles, about anything you want to know about what's going on on the Earth in real time every day. The, the number of use cases of that is gigantic. And, um, I think we are unlocking that at a pace that was So we were going a bit slow on some of that analytics, and I think it was hard for those companies that were just in the analytics space for that reason. But now we're at takeoff.

And you guys have more images than probably anyone else. Yeah. And so in, you mentioned the farming example, like what's another one that's blown you away?

So what another application of AI, let me give two Well, uh, one in, uh, sustainability and with civil government. So we work with the Brazilian Federal Police. We monitor 8 million square kilometers of the Brazilian Amazon. Their problem is how can I detect where the deforestation is? So we already imaging all of the Brazilian Amazon. We take that, use AI to look for new roads where it finds new roads. That's a sign of illegal deforestation or illegal narcotics or illegal mining. We find all three. They then send people out to, um, to, to stop that. They did 3,000 interventions last year, confiscating $3 billion worth of assets. And sometimes they showed it, sent us videos of them putting dynamite on them and blowing them up. And most of all, they reduced deforestation rates by, uh, more than half. So they halved the deforestation rate.

The dynamite is on the mine and not some person?

I'm on the machinery that they confiscated. Yeah, the problem is because if they just took it to the local police, then it went back round again. So they decided to take a different tactic.

Get rid of it.

Dynamite them and then they won't come back round again. That's one strategy. Another one is in the security field. And obviously federal police has sort of got some security implications, tracking narcotics and other things. But in the deeper security space, we've been working closely with Ukraine and helping them with understanding threats deep into Russia, understanding, giving warning of those things really early. So we image all of Ukraine, all of Belarus, all of Russia every day, and they track, they use AI to find changes, and not just of military capabilities, but of industrial capabilities that might be supplying a supply chain that they want to stop. Up, right? And, um, they do battle damage assessment with— whether it's the Russians have hit them or they hit the Russians, what has happened, has it succeeded or not. They do it for planning their drone flights. Their drones are, um, half of their drones weren't coming back when they sent— they have these ISR drones, so imagery drones that went up to go battle to understand what was going on, but half of them were being shot down and not coming back. So instead, now they just look in our imagery and if there's no changes, they don't need to send that drone, saving them billions of dollars. But integrated into all of this is AI, because it's a huge area. Um, they're monitoring 18 million square kilometers every day, just like Brazil is doing, but this time for security threats, not for deforestation. And they're alerting themselves. So whilst Ukraine has a 7-to-1 tank and troop disadvantage to Russia, they've had a massive information advantage because of satellites and AI. And so I think the security implications of satellites and AI are huge. And we've always said our main mission is sustainability and security, and those things are increasingly coupled.

Yeah, I mean, I wrote about this in the book and, and have written about it subsequently. I mean, I would argue the Russia-Ukraine war was our first space war in, in some ways. From the very first— well, before the war even kicked off, it was often Planet's imagery that was appearing on the page of the New York Times or on CNN of the Russian troops amassing in Belarus. And Putin said he wasn't going to attack.

And what's all this then?

Yeah. And then we had, we had photos. Obviously Starlink has been used a lot by Ukraine. Yeah. Do you think, do you think the— so, and again, this is not just because you're sitting here. I mean, I argue this stuff all the time. I mean, I think people underestimate how valuable this technology has been. You can make a pretty pretty convincing case, I think, that— I mean, who knows how the simulation would play out, but that Kyiv would have— would likely have fallen without some of these, um, technological assets in the, the early days of the war. Do you— no one would know this better really than you. I mean, do you think the Russians were caught off guard by— obviously they knew this technology existed. Do you think they were just caught off guard by how it would be used. The fact that, that all this— when their troops were getting stuck on the way to Kyiv, I mean, it was just a massive morale-busting, um, yeah, to see these, these images that, that no government could spin because they were coming from a neutral party. And, and it just immediately took away— not that the Russian army isn't scary, but, but this thought of this, this unstoppable monster going up against this, you know, much smaller army. I mean, it was just like, God, these guys look kind of incompetent. I just feel like it really changed things. And then not to mention, then they would go on to, "We didn't bomb a hospital." And then you have a picture of, "Well, it was a hospital yesterday." All these things for their propaganda campaign. Yeah, do you think they were caught off guard by this?

I do. I think they were caught off guard by two things. One is the, um, the fact that Ukraine— Ukrainians thought they had a country and Russia thought that they didn't, right? They actually believed in their country.

Yeah.

And, uh, so I think bizarrely enough, they didn't think they would, and, uh, they got caught off guard by the technology and innovation. And obviously necessity is the mother of invention, so Ukraine was being very deft in leveraging these new tools and applying obviously with support from the West, intelligence and other things that has been— but yeah, I think that early on it made a very big difference to the success or otherwise of Russia in just taking Ukraine in the first place, and subsequently in holding accountability, um, across the boards. We've also helped all the NGOs that have been there trying to supply medicines, the World Food Program in trying to get grain out of Ukraine because half poorest countries in the world were being fed by grain from Ukraine. It's a breadbasket, and literally half of the World Food Programme's grain was coming from Ukraine. So helping them get it out, various other humanitarian things. But I think one of the most understated things is just how everyone could see what was going on in real time. And I think not only did they make a mistake in that, but the question now on everyone's mind, I think, is how does it— what implications does it have going forward for geopolitics? Now countries know that everything is going to be viewable to everyone, so they can't just make shit up. And Putin may not have guessed that that was going to be exposed so readily, but, uh, you know, future countries might now know that that's what's going to happen. And so, uh, and they can shield some of that from their own, uh, people, as Russia has done, but it's hard when you're in hearts and minds campaign for the whole world. And everyone gets to see these images, and these images don't lie. And so, yeah, I think it played an important role there, an unstated role. I don't think it was the first space war, by the way. I think Iraq was the first space war, but that was all government. Right. This is the first commercial one. And it has a—

I mean, the Cold War might be the first space war.

In a way, yeah. But Iraq was where space was integrated into their military operations, comsats, GPS, guided missiles, all this. But this was the first time everyone else got to use it.

That's what I mean. I think of it as this, where the— you actually had commercial technology that was, was better than what the— at least better in some cases, and quite different than what the military had.

Yeah, absolutely. And various organizations were using it, you know, that NGO, the, the, the people looking for war crimes, the, um, the all the other countries that are trying to support Ukraine, Ukraine itself, a common operating picture between them. The news media, the think tanks shedding light on what's going on, open-source intelligence communities that are triangulating our data, cell phone data, this, to find out what's going on and get to the truth of things. What a— and in a world where we see more and more misinformation online, uh, driven by algorithms and incentive structures that are misaligned, here we have a situation where we can triangulate truth to bring that truth and get back onto a common operating picture. I I think an underappreciated point about planet— people think about planet and saving the forests and stuff. One underappreciated point is how much role we can play in security and how important that is, that transparency and accountability. Another thing is the implications for everyone to have a more clear, common operating picture of what's going on. And we have lost a little bit of that. You know, we can't agree whether climate change is real, who won the last presidential election, Do vaccines work? Well, here's a way in which we can get back onto the common picture, literally a common picture of what's happening. And you can argue about interpretations of the picture, but in the end, the picture's saying what the picture's saying.

But then, since I've known you, I mean, you've always been, I think, on the side of more information, I would say on the side of truth and justice and knowledge and things like that. But you, and I I, you know, I've written about this, but I don't— a lot of people probably don't know, you still, because of this, you can end up getting caught between opposing forces who may or may not want your images out there. And whether you do Russia-Ukraine, whether you do Israel, um, I feel like this last few years must have been quite interesting for you in terms of the phone calls that you're getting. And, and it seems It seems like sort of awful to be the— in some cases, I imagine the decision goes up to you and to be this like arbiter on what the public gets to know or not know.

Well, there's a lot, a lot of it, and it's very sensitive and we get yanked into these world events. I mean, you know, I can be on a weekend, but if Iran-Israel is erupting, you know, it will definitely pull us in. We're sort of involved directly. And I think it's critical that we err on the side of giving the data out to all the parties, because then it's less likely that they will make mistakes. History is full of examples where people made assumptions or mistakes because of a lack of information, and then that led to more escalation. Into war. I think the more we have a transparency of what's going on, the less likely that is. Economically, in the Cold War, there's a— Russians put in missiles in Cuba, almost led to it going hot. Or vice versa, when the US put missiles in Turkey that the Russians didn't know about, that it almost got hot. It was not when they knew where each other's stuff was at. When they knew where each other's stuff was at, it was more peaceful. And I think that lesson, is what we took into our theory of change on the security side for Planet. So our thinking is we don't always give to all people, uh, terrorist organizations, um, North Korea and the likes of countries that are embargoed under U.S. law and so on. But on the— on generally, we will give it to anyone that will pay for period. And but we won't be doing that exclusively. So if Reuters wants it, Reuters can have it. If the UN wants it, they can have it. If Israel wants it, they can have it. And everyone can triangulate what's going on. And we think that's a good thing.

But you— and okay, I'm going to give you a couple tough questions now.

Uh, go right ahead. Okay, you know, I've had them all before.

The US— you have? You have? Okay. And sometimes I find your answers satisfying and sometimes not. So let's, let's find out. I mean, the US government is one of your biggest customers. So how can somebody trust when you say some of these things? Because obviously they might come to you on some occasions and be like, please don't put that picture out. And, and you're on the side of wanting this information to be out there, but you've, you've, they're one of your biggest customers on, on many different fronts. Yeah. I mean, how, um, This must be a very hard calculus to go through.

For sure. Well, let me say, um, the US government has never asked us not to give an image out.

Okay.

And they have almost always defended our ability to provide imagery when other countries have put pressure on us. So we have never stopped doing that. Um, look, we are very proud of partnering with the US government for all the reasons that, that it, I think, is an empowering tool for all countries to have greater understanding of security We work with the NRO, we work with NASA, we work with Space Force, and truth is power. And I think that's really important. And as we've seen in Ukraine, information advantage can be critical. And I'm on the side of more open societies, ones— and this data is less threatening to more open societies than to more closed ones. And I think that's a good thing. Does everyone agree with that? No, but that's where we stand.

Well, and when I was doing the movie, I mean, the biggest question that I got from people was, okay, and we talked about this a little bit, but you were, you know, I call you space hippies in the book, which I know you guys don't always love, but I mean it.

But then you put the title Heaven's Content.

I mean it. Funny, isn't it? In a loving way. I mean, you were, you as a youngster, you were fighting the militarization of space. You were part of a group group of young idealistic kids. You started Planet. I don't even think it was really meant to fully be like a for-profit enterprise at the beginning. I mean, the very foundation of it was to, to understand the Earth. And, and obviously, look, I mean, you end up running a public company. This business needs money to grow and survive. You have deals with all kinds of governments. Um, I mean, this was the question I got over and over again, especially on you from other people was, was kind of like, how do you rationalize that? And you would always tell me, you, I think you were more of like a scientist than a capitalist, or, uh, you know, and, and have, I would say, um, why, uh, you can, you can say how you characterize it, but sort of socialist, um, monetary philosophies, you know, around your own life. So I, I've, I have always wondered how you square all these things in your head? Because you are a, you're a, you're super bright, thoughtful person, and it seems like an impossible number of conflicts to me.

Yeah. Let me try and be as clear as I can on the security one first. We are enabling transparency that enables accountability on the earth that I think will lead to more peace and global peace and prosperity. And on sustainability, the connection is obvious, you know, you can't manage what you don't measure, track all the trees, check all the coral reefs. And there's huge business value, there's an agricultural insurance, finance and all that. And that is a tool that enables us to scale more more. I personally think our mission is more important than the business. I think you'd find that for a lot of companies. You know, Google here wasn't started as a— focused on the business and revenue. It was focused on access to information, right? We have that bigger purpose, you know. We're not Google or Google scale, but we have that missionary purpose as more important than anything else, and I'll continue to stand by that. But the security piece fits in, and security and sustainability are completely coupled these days. Um, security situations lead to sustainability ones and vice versa. Um, uh, I already just gave you the example in Ukraine. Um, because of the war, the food wasn't getting out to the people that needed it, so it was leading to a, a food security issue around the world. You have other examples like Syria Syria had a drought, um, because of climate that led to a crop failure that led to civil unrest that led to refugees in the neighboring countries and in Europe. Um, those things are coupled. The Pentagon's top threat, it often talks about the top few threats to the United States, always, no matter your ideology, um, at the top and your political leanings, it, because it is fully rationalized about and looking at these things puts climate as one of the top threats. Why? Because droughts and, and heat waves and what have you lead to more unrest that it then has to worry about in terms of threats around the world. So climate and environment and security are totally interlinked, and that's why to me it's not a one one or the other or reconciling one and we're getting military contracts so that we can do impact work over here on deforestation. Not at all. It is coupled. I mean, the Brazil case kind of gives you that sense, right? I mean, they're stopping illegal action and stopping deforestation and it's all there. That's how I see Planet fitting in. And vis-à-vis the capitalist side of it, Chris Kemp often tweaks me on this because he's like, you're excellent capitalist, see?

And I'm like, you're better capitalist than Chris is at the moment.

Well, he does find that ironic sometimes. We, um, for me, I want to lean in to the engine of growth to enable this mission.

But if you're anti- if your personal leanings are to kind of question, you know, some of these systems.

My personal leanings is to question almost everything. Have you met me? Like, come on.

Yes. But then you, you end up, you know, I, I do not, okay. So people should know, I, I've known you for a long time. Sure. And yeah, I mean, I, I would not, if I'm being honest, I would not have seen you as like a public company CEO of, you know what I mean? You, you were, I always thought of you like, oh, this is kind of like a do-gooder kind of guy, you know? And not that you— again, we've been over a bunch of reasons how you are. I think I also thought you might have just like lost interest in like, you know, running a business, doing all this, losing interest in saving the planet.

No, no.

But you're so— you were like such a scientist at heart. And, and like, I just— I would imagine— well, now I am a CEO, so I know. I mean, there's things that come with doing the job that is, you know, is pretty far from what I want to be doing, which is writing stories and making videos and things like that.

Well, sometimes I miss being tweaking satellite stuff in the lab, but honestly, it is offset by the impact that we are having, whether it's in Ukraine or in Brazil, as the examples we've been talking about, or that farmer, um, uh, every day. And the ability for me to help enable hundreds of people at Planet to do that mission by raising capital or by helping to make connections that enable that partnership with NATO or what have you is fulfilling because of the outcome. And more than just enabling those people at Planet, enabling thousands of customers to leverage the benefit of that information. Now, I want it to be hundreds of thousands of customers and millions of customers because I believe we're only 1% or less of our journey in terms of business and in terms of impact, and they are aligned, uh, in our case. And so, uh, I'm not going to stop until we've scaled this. And I feel very good, uh, about our business right now. We've got a lot of momentum on our side, and I'm leaning in. AI is an accelerant and all these things.

And you say you don't have sleepless nights in your communal house?

I sometimes have sleepless nice, but not often. Not often. I'm a pretty good sleeper, actually.

Why has, like, as far as I'm aware, I mean, there's obviously other imaging companies, but has like China built something that's equivalent or roughly?

No one's doing a daily scan as yet. And the other imaging companies in the West are all doing less than 1% of the landmass of the Earth. So people often go, oh, there's Maxar, Airbus. They're all doing very different things. If you look under the hood, um, it's not like comparing, uh, uh, uh, uh, two electric car companies. We are scanning the whole Earth, which opens up agriculture, which is 30% of the landmass of the Earth. And so you can't do it if you're only imaging 1% of the landmass of the Earth, like Maxar, Airbus, or Black Sky, all these guys. They have an application That's very cool. And we do some of that too with our high-res, but it's actually quite different applications. So I don't see it as particularly competitive. The Chinese have started to build a system somewhat similar to us called Jilin, which hasn't yet done a daily scan, but is impressive what they're building. And it's good to get the competitive juices going.

Yeah. Well, and I know, look, you guys do many sophisticated things and this is not easy. Are you surprised that there hasn't been a more direct competitor?

Not really. It's really hard. I mean, you know, as I like to say, uh, PowerPoint satellites are easy and actual satellites are really hard. So not many organizations can actually pull it off, um, to launch and operate not just one satellite but actually make it work, to then do hundreds of satellites, to the mission control systems, the ground station systems, the data processing systems, It is not for the faint of heart. And so most of them fall over before they get to that scale. We've got an incredible team that has incredible skill sets in this department, starting with a core team that we took from NASA. And I was so proud that 10 years in, when we celebrated our 10th anniversary a couple of years ago, 7 of the 10 first employees at Planet were still at Planet. That institutional knowledge we have. And some people leave and we call them boomerangas because they almost always come straight back. They go to SpaceX, they go to Google, and then they realize that it's better at Planet.

I will attest to some of that. It's a good company culture. I'm looking at how much time I've taken up. There's still a few things I was hoping to get to.

Yeah, you okay?

Yeah. Do you, Why did you guys— again, this is a question I've never asked you, even though I probably should have. Do you regret not doing communications alongside all of this? You'd be like a multi, multi, multi, multi-billion dollar business if you beat SpaceX.

Maybe if we were to pull it off. I mean, one of the things I give credit to SpaceX for, their execution on Starlink. I think I was the first non-SpaceX user of Starlink because I have a rural property and I needed internet, and so I, I called a friend. I called Elon and got him to send me one. I'm pretty sure before he'd sent any others outside of SpaceX. And I— it's impressive what they have. And this is way before there was an operational system. You could occasionally get internet access. Um, but, uh, that's a hard job. And one of the things that's really hard about it is getting to the scale that gets to global coverage. Consistently, because it's kind of not useful if you can only get 3 minutes a day of uptime, right, in your rural property. I mean, maybe it's useful for some applications, but generally you want communications on all the time. So until you get enough satellites to do the global coverage, there's of order 1,000 satellites that we didn't have the capital to do, right? Whereas in Earth imaging, if you get a couple of satellites up, you can already start selling that information. Yeah. Okay. So the entry barrier was much lower. Then on the other side, I think the applications are much stronger. I mean, not to take away from the applications in communications satellites, teleeducation, telemedicine, all these fantastic things. Um, watching Netflix, watching Netflix. Um, then on Earth observation, look at the challenges around the world. Um, you know, we are pulled into sustainability, security, um, that we're relevant to all the things to do with food security, water security. Look, the UN has the Sustainable Development Goals, basically a list of the world's biggest challenges, poverty and so on. It's— we estimated that we can help 14 of those 17 goals in terms of measuring it. And if you don't measure it, you can't really achieve it. And 140 of the 200-something sub-indicators under those 17 goals. I mean, our data set is in principle relevant to so many of the world's challenges. So I think the impact mission of Planet is much bigger than communication satellites will ever be.

You could probably— this is what I feel though, is ever since the ZEI race took off, people seem to talk about climate change a lot less all of a sudden. In the last election, it wasn't even a big deal, and then now I think we've all given over to this idea that, well, this is the next big thing. We gotta race there. It's gonna take a lot of energy. Let's, uh, there's a race dynamic in AI, you know, and then it's always, it's like, well, well, but then the AI's gonna figure out climate change, so it's gonna fix it. Let's just, let's just get there first.

Yeah. I mean, that race dynamic is obviously present between the companies and between countries, between the US and China. And there's, you know, and, um, but, um, let me just talk about how it's relevant for Planet for a minute because it's just unlocking everything. As I, I already sort of mentioned, it's accelerating our acquirable Earth vision of indexing what's on the Earth and making it searchable. But the speed at which that is happening, the basic ways in which that helps our users is helps them speed time to value, go from zero to an answer. If you're that farmer that I sort of had you imagine earlier, you don't want to look at all the images. You just want to type in a question. Hey, What should I do to improve my crop yield? Oh, you should look at this area of that field and add some fertilizer or what have you. Or there's blight, cordon it off, whatever it is, the action. AI is enabling it. So speeding time to value. And then there's democratizing it. It's the other way. Because whilst it was just the big companies, we work with Bayer as an agriculture company. They look at millions of farmers' fields every day with our data. But that's all very well for them, big ag company, um, how they have a geospatial team. But what about that individual farmer that doesn't have that geospatial team? So it's democratizing. So it's speeding time to value and it's democratizing. If I put a slightly bigger philosophical picture together for you, again, it's the— all these AIs, ChatGPT and so on, are just talking about the text of the internet. And here we are, have the potential to now be able to chat interface with the Earth and everything on the Earth in the physical world. And bringing that physical way layer into the chat bots is an— will enable huge applications across sustainability, security, digital transformation, and so on that I think are vast. So, um, also I'd like to point out that, um, it's funny, a lot of people left Silicon Valley because they were like, oh, it's going to be better in LA and Texas and Florida. And they've all come back with their tail between their legs because all the coolest shit is happening here.

Yeah.

And we are in the middle of what I call the AI triangle. We're a few blocks from the Anthropic team, which we've announced a partnership with. We're a few blocks from Google Gemini team, which is not actually down here in Mountain View. It's the Gemini team doing that cool stuff is all up in San Francisco and a few blocks from, uh, uh, the OpenAI team. And so we feel in the thick of it, and we are in the thick of this revolution that's happening in AI that enables the, the, the extraction of value from our dataset. Our dataset is uniquely able to do that.

Uh, no, and I, I appreciate everything you're saying about the technology, where it's going. I just, yeah, I get a little depressed because I feel like some of the attention— we had this like peak for a while when people were super into climate change, made all this money going into green tech companies. And it just feels to me like a lot of the things that you care about.

I think, as I said, AI companies are going to have to proactively steer that technology. Otherwise, their default path will be to accelerate our blowing up of the planetary boundaries. Obviously, I feel like we're part of that. I think we're of it, we can help make AI safer and we can help use AI to save life. So, like, if we want AI to care about life, including humans, we want it to know about that stuff. So we don't want to just train it on the text of the internet, have it tell us about Wikipedia. We want it to know about all the trees and the corals and all the living things. So we better start feeding in that information. So So we've got our own existential interests in training it about that. And then there's business interest because obviously you want the real— if you're trying to tackle real-world problems with AI, you want real-world data. If you want to do an insurance claim or respond to a disaster, you're not going to do that with fake data. You need real data. You can't just imagine that data. And then so, so, so for all these reasons, I see a confluence of how AI, what AI is doing, is going to need our, is it going to need our data to both for business reasons, is it to have a better understanding of the world and for environmental reasons to have, you know, if we're going to tackle climate change, it needs all the data about climate change, right? And they're going to have to steer it actively towards that if we're going to have a good future here.

Yeah. Okay. We're going to go a little rapid fire around, right? It doesn't have to be, but it could be. And some of these is a bit random. So, you know, you, you not only predicted but helped usher in this massive breakthrough with the, the price performance of satellites. As I write, that was about in the book. Yeah, you know, you, you kind of brought Moore's Law to space. It's, it's in many ways more dramatic than we've even seen with the fall of, of rocket launches and the price of getting something to space.

Yeah, it's 1000x over 4x for the rockets. Yeah.

Yeah. And there's— okay, so there's millions of wacky ideas around where space is heading and what's going on. I feel like you, you're such a good person to ask given your, your history. I mean, what's something that like most people wouldn't think about that you think will actually happen in space in the next— I mean, you can pick your time period, 20 to see years.

Dirty infrastructure into space.

Yeah, putting, putting—

yeah, energy production, compute, uh, you know, I think that putting server farms in space is a no-brainer once you get to a certain launch cost per kilogram. You don't have to clean the solar panels and cool it. You, you just do that automatically and have these data centers upstairs, um, automatically online. That didn't make sense made sense a few years ago. It still doesn't quite make sense, but it is on the verge of. I think that's important because we can take that dirty infrastructure off Earth, less pressure on our ecosystems. So it's a win in lots of different ways, as an example. So I see, I mean, we're already going in that direction. I mean, this launch we have in a couple of weeks here from Vandenberg, we have the latest NVIDIA chips on it, with, uh, uh, the ability to do AI processing at the edge. We've done this partnership with NVIDIA. You just imagine having more compute at the edge to do that processing there. That means just like, um, recognizing those ships right there and then as soon as you take the picture, and then sending down the latitude and longitude of the ships to the people that really need that, rather than waiting for it to download the whole data, which might take longer over ground station or what have you. We're going to see That will speed up, make our insights more real-time. And it's sort of a stepping stone towards putting the compute itself upstairs.

And then I know you care about this because you used to work on space debris at NASA. Can you give me an answer better than like, I kind of don't know how this is going to play out? I mean, as far— as we're sending tens of thousands of satellites to space, what you just talked about, we're going to send even more stuff. I know you worry about this type of thing. Don't see— I know there's some stuff in Japan and a little thing here and there. Don't see a lot of people working on space debris, on trying to avoid these things running into each other and what would happen. I don't know. I mean, what's your current feeling on this? And any reason for optimism? Well, yeah, I think there's—

yeah, the reason reason for optimism is that actually relatively simple things can ensure that we, um, don't cause a problem, like keeping the satellites lower where the atmosphere of the Earth just self-cleans those orbits. I mean, they still re-enter the top of the atmosphere, but we're still talking tiny amounts of actual mass compared with other pollutants and other things we put in the atmosphere, and it's mostly, uh, inert substances. But, um, and don't worry, it won't fall on your head because they naturally survive re-entry, so they just burn up like a shooting Star. That keeping stuff low is the way forward. We— the bigger problem is not satellites, uh, the, but the debris, um, which are the leftover bits of satellites when satellites blow up and things. And their, their practices like draining the tank before you end life the satellite, it means that there's not a— if there's a spark, it doesn't blow up, right? We've been doing that for decades now, but there's a lot of stuff still there. Um, I think there's two big— two or three big things we've got to do going forward. One is we've got to ban anti-satellite technology, that is the deliberate kinetic blowing up, which some countries have been doing, which China, the US, India, and others have done in recent years, which causes bigger mess. I mean, you, you— in one go you make 10,000 pieces of, uh, of debris that you can't even track because it's so small, and that, that It'd be catastrophic to collide with other satellites. So we've got to stop that. It's asymmetric environment in space, not like on the land, sea, and air where collisions— the debris actually can be collected. In space, it can't. It causes a long-term problem. So we have to think differently about kinetic action in space, and I think it should just be banned outright by all the countries. The second thing is megaconstellations need to have not just stay— they need to stay low, not just have propulsion, because propulsion, the problem is even if 1 in 100 of the satellites, a propulsion system fails, and even if you have redundancies, you still have a lot of dead mass up in orbit, so you can't deorbit it at the end of life. So I think we've got to stay way low. And, and then, and then the third thing, um, ultimately we're going to need some sort of traffic management, uh, system, um, and debris removal services. And you said there's not many people working on that, so it's right, but there are ways uh, one scheme, one of the few good ideas myself and a few others came up with at NASA, was a laser scheme for nudging space debris. So you propagate forward all the positions of the, the known pieces of debris. If two are going to collide, you nudge one using light pressure. So the light just nudges it so that it misses the other piece, and then it stops the cascade. With one laser system in one telescope in the, uh, pole you need in Antarctica or the Arctic, um, you can nudge most and be sort of like a traffic cop for the debris.

So there's a laser at the pole?

No, there needs to be.

No, no, but you would have a laser at the pole and it would, it would be monitoring what's going on with these radars around the world, and then, and then it would just blast—

it wouldn't blast it, it's not like ablating it, it's literally photon pressure so that they miss each other. So if you see two that are going to collide, you just nudge it. It's ever so slight, but you just wanna stop those places where it actually collides because that, uh, that is the form, the beginning of the Kessler syndrome, right? Is, is the collisional cascade that you're trying to avoid. But with a little bit of nudging, you can be a, a traffic cop that stops the collisional cascade. So I think that's the kind of thing we need to do in order to have traffic management long term.

You're friends with lots of very rich people. Have you ever pitched them on this idea?

This is actually, I pitched NASA on it and unfortunately Originally it was too, too cheap. You know, NASA doesn't know what to do with projects that are under $1 billion. So this was only going to cost $10 or $20 million. And I was like, gosh, but there's some, some really interesting projects on it subsequently, including by the Australians that have been testing a laser through a telescope system there to nudge debris. And I haven't caught up on the latest, but I think that's the system that needs to happen. We've got a little bit of time because it's been the very early part of the Exponent prominent on the Kessler syndrome, but we need to get on it in the coming years. All right.

I promise to end on a positive question right before that. You worked at NASA for a long time. You were quite controversial while you were there. You mentioned they don't love to do cheap projects. That was a problem you kept running into. You did a low-cost lunar lander that you had to hide because it was so expensive.

And then there was the phones in space.

Right.

Almost got fired.

And I know the US government is a big customer and we all, we all love NASA in our heart.

I—

but quite easy to argue, seen better days in many ways.

Are you—

are you like, what should NASA be doing? What would you change to NASA right now if you, if you were the administrator?

Well, great question. I mean, I would— I think that that we could do more science for an exploration for less money.

Like, let's pretend right now you're the administrator.

Yeah.

And you can do— they have to say yes to like whatever the— to, to 2 or 3 things that you— they just have.

Well, I'm biased, but I would lean it more into commercial thing. Take this Tanager mission I just mentioned, um, the hyperspectral satellite. That was this collaboration between us and JPL. It was going to cost $300 million per spacecraft to do that in their Earth systems budget. Um, now we can pump out that spacecraft for about $6 or $7 million including launch. It took us some real R&D to get there, but now it's tens of millions of dollars, but then now we can pump them out. So we can launch a fleet of them, vastly lower cost. So the collaboration between NASA's exquisite instruments and commercials, uh, making things cheaper and more affordable at scale, scale can allow you not just to do the same thing, but like to do the same thing with less money and faster. So I would apply that general rule across a wide variety. Now there's going to be some things that NASA, you know, can't do that way. And I'm protective of NASA because I, you know, I used to work there and I feel some of the recent decisions have been, what do they say, um, ready, fire, aim. You know, it's been decisions before thinking through how to do it in a smart way. The intent has been very good, I think. How do we figure out how to do this cheaper? There's been a revolution in space, in rockets, in satellites. How do we do this cheaper? The intent, I think, makes a lot of sense. We've just got to do it with a little bit more precision. And I I think companies like Planet stand ready to really help with that if we can.

That wasn't the ruthless NASA administrator answer that I was expecting. Okay, last one. Is there one particular image, you know, you've pulled down from your satellites that is the most memorable to you?

Well, it's probably our very first image. I can see it still in my eye that it was of area in Oregon with forests and agricultural land. And just to see the resolution, because you can, you know, that you believe in the physics equations, if you— there's this telescope size at this distance in orbit, we should be able to see this level size object on the ground. But until you actually fly, you don't know for sure. When that image came down, it was huge for us, and it showed that everything doing then was just about scaling, just about scaling. That's still hard, but like it was just that. And so yeah, I still remember that very first image.

Well, that makes, I mean, that makes a ton of sense. People can watch the movie, read the book for some of the backstory, but was born out of this idea that began just sending cell phones up into space on the hunch that maybe consumer electronics would work in space and that we'd be able to do spacecraft this 1000x-ish cheaper.

You said it at the beginning, mainframes to desktop computer sort of transition for the space industry. And it's, that still hasn't taken off even as much as I thought it would, but we are now really in the throes of, so if it wasn't for the AI revolution, people would be like, whoa, there's a big space revolution happening. I mean, some people obviously track it, but on the scale of the globe. And my point, most of all, is that these two things are colliding, the space revolution and the AI revolution because they're natural partners.

Well, it's all very dramatic. And if, like you said, I mean, if we hadn't been putting more modern computing into space and on some of these systems, we wouldn't even be able to get there. And I always think you guys, I just don't think it's like unreasonable to say, I don't even know, this definitely doesn't, SpaceX gets a lot of credit for a lot of things, but I, like you guys were the ones that had this vision of pushing modern electronics and computing into space and really changing what we could do in low Earth orbit. And I don't know, obviously I care because I did all this stuff, but sometimes I think you guys don't get the credit you kind of deserve writ large for what is just like a monumentally dramatic shift.

Which we still haven't seen the full implications of. Well, thank you for for drawing attention to it. I appreciate you, uh, seeing that.

I try, I try. Um, Will Marshall, thanks so much for coming on the show. I appreciate it.

No worries.

Thanks. The Core Memory Podcast is hosted by me, Ashlee Vance. It is produced by David Nicholson and me. Our theme song is by James Mercer and Jon Sortland. And the show is edited by the John Sortland. Thanks as always to Brex and E1 Adventures for making this possible. Please visit our Substack, YouTube, and podcast channels to get more of what Core Memory makes. Thanks, y'all.