The following remarks were made on May 9, 2019, in Washington, DC, at an event to unveil Blue Origin’s lunar lander, Blue Moon.

BLUE ORIGIN IS the most important work I’m doing. I have great conviction about it, based on a simple argument: Earth is the best planet.

The big question we need to ponder is: Why do we need to go to space? My answer is different from the common “plan B” argument: the Earth gets destroyed and you want to be somewhere else. It’s unmotivating and doesn’t work for me. When I was in high school I wrote, “The earth is finite, and if the world economy and population are to keep expanding, space is the only way to go.” I still believe this.

The question “What’s the best planet in this solar system” is easy to answer because we have sent robotic probes to all the other ones. Some inspections have been flybys, but we’ve examined them all. Earth is the best planet—it is not close. This one is really good. My friends who want to move to Mars? I say, “Do me a favor. Go live on the top of Mount Everest for a year first and see if you like it—because it’s a garden paradise compared to Mars.” Don’t even get me started on Venus.

Look at Earth. It is incredible. Jim Lovell, one of my real heroes, while he was circling around the moon on the Apollo 8 mission, did something amazing. He put out his thumb and realized that, with it at arm’s length, he could cover the whole Earth. Everything he’d ever known, he could cover with his thumb, and he said something amazing. You know the old saying “I hope I go to heaven when I die.” He said, “I realized at that moment, you go to heaven when you’re born.” Earth is heaven.

The astronomer Carl Sagan was so poetic: “On that blue dot, that’s where everyone you know, and everyone you ever heard of, and every human being who ever lived, lived out their lives. A very small stage in a great cosmic arena.” For all of human history the Earth has felt big to us, and actually in a really correct sense, it has been big. Humanity has been small. That’s not true anymore. The Earth is no longer big. Humanity is big. Earth seems big to us, but it’s finite. We have to realize that there are immediate problems, things that we need to work on, and we are working on those things. They’re urgent. I’m talking about poverty, hunger, homelessness, pollution, overfishing in the oceans. The list of immediate problems is very long, and we need to work on those things urgently, in the here and now. But there are also long-range problems: we need to work on them too, and they take a long time to solve. You can’t wait until the long-range problems are urgent to work on them. We can do both. We can work on the problems in the here and now, and we can get started on the long-range problems.

We want to go to space to protect this planet. That’s why the company is named Blue Origin—for the blue planet, which is where we’re from. But we don’t want to face a civilization of stasis, and that is the real issue if we just stay on this planet—that’s the long-term issue.

A very fundamental long-range problem is that we will run out of energy on Earth. This is just arithmetic. It’s going to happen. As animals, humans use ninety-seven watts of power—that’s our metabolic rate as animals—but as members of the developed world, we use ten thousand watts of power. And we get a lot of benefit from it. We live in an era of dynamism and growth. You live a better life than your grandparents did, and your grandparents lived better lives than their grandparents did, and a big part of that is the abundance of energy we have been able to harvest and use to our benefit. There are many good things that happen when we use energy. When you go to the hospital, you’re using a lot of energy. Medical equipment, transportation, the kinds of entertainments that we enjoy, the medications we use—all these things require a tremendous amount of energy. We don’t want to stop using energy. But our use levels are unsustainable.

The historic compounding rate of global energy usage is 3 percent a year. It doesn’t sound like very much, but over many years the compounding becomes extreme. Three percent compounded annually is the equivalent of doubling human energy use every twenty-five years. If you take global energy use today, you can power everything by covering Nevada in solar cells. Now, that seems challenging, but it also seems possible, and it is mostly desert anyway. But in just a couple hundred years, at that 3 percent historic compounding rate, we’ll need to cover the entire surface of the Earth in solar cells. Now, that’s not going to happen. That’s a very impractical solution, and we can be sure it won’t work. So what can we do?

Well, one thing we can do is focus on efficiency, and that is a good idea. The problem, though, is that it’s already assumed. As we’ve been growing our energy usage 3 percent a year for centuries, we have always focused on efficiency. Let me give you some examples. Two hundred years ago you had to work eighty-four hours to afford one hour of artificial light. Today you have to work 1.5 seconds to afford an hour of artificial light. We’ve moved from candles to oil lamps to incandescent bulbs to LEDs and gotten tremendous efficiency gains. Another example is air transportation. In the half century of commercial aviation, we’ve seen a fourfold efficiency gain. Half a century ago it took 109 gallons of fuel to fly one person across the country. Today, in a modern 787, it takes only 24. It’s an incredible improvement. It’s very dramatic.

How about computation? Computational efficiency has increased one trillion times. The Univac could do fifteen calculations with one kilowatt second of energy. A modern processor can do seventeen trillion calculations with one kilowatt second of energy. Now, what happens when we get very efficient? We use more of these things. Artificial light has gotten very inexpensive, so we use a lot of it. Air transport has gotten very inexpensive, so we use a lot of it. Computation has gotten very inexpensive, so we even have SnapChat.

We have an ever-increasing demand for energy. And even in the face of increasing efficiency, we will be using more and more energy. That 3 percent compound growth rate already assumes great efficiency gains in the future. What happens when unlimited demand meets finite resources? The answer is incredibly simple: rationing. That’s the path we would find ourselves on, and that path would lead, for the first time, to your grandchildren and their grandchildren having worse lives than you. That’s a bad path.

There’s good news: if we move out into the solar system, we will have, for all practical purposes, unlimited resources. So we get to choose: Do we want stasis and rationing? Or do we want dynamism and growth? This is an easy choice. We know what we want. We just need to get busy. We could have a trillion humans in the solar system, which means we’d have a thousand Mozarts and a thousand Einsteins. This would be an incredible civilization.

What could this future look like? Where would a trillion humans live? Gerard O’Neill, a professor of physics at Princeton University, looked at this question very carefully, and he asked a very precise question that no one had ever asked before: “Is a planetary surface the best place for humans to expand into the solar system?” He and his students set to work on answering that question, and they came to a very surprising, for them, counterintuitive answer: no. Why not? Well, they came up with a bunch of problems. One is that other planetary surfaces aren’t that big. You’re talking about maybe a doubling at best, which is not that much. And they’re a long way away. Round-trip times to Mars are on the order of years, and launch opportunities to Mars occur only once every twenty-six months, which is a very significant logistics problem. And finally, you’re far enough away so that you’re not going to be able to do real-time communications with Earth. You’re going to be limited by a speed-of-light lag.

Most fundamentally, these other planetary surfaces do not and cannot have Earth-normal gravity. You’re going to be stuck with whatever gravitational field they have. In the case of Mars, that’s one-third G. So, instead, O’Neill and his students came up with the idea of manufactured worlds, rotated to create artificial gravity with centrifugal force. These are very large structures, miles on end, and they hold a million people or more each.

A space colony would be very different from the International Space Station. Inside it would have high-speed transport, agricultural areas, cities. The stations don’t all have to have the same gravity. You could have a recreational colony that kept zero G so you could go flying with your own wings. Some would be national parks. These would be really pleasant places to live. Some of these O’Neill colonies might choose to replicate Earth cities. They might pick historical cities and mimic them in some way. There would be whole new kinds of architecture. These are ideal climates. This is Maui on its best day all year long—no rain, no storms, no earthquakes.

What does the architecture even look like when it no longer has the primary purpose of shelter? We’ll find out. But these colonies will be beautiful—people are going to want to live there—and they can be close to Earth so that you can return, which is important because people are going to want to return to Earth. They’re not going to want to leave Earth forever. They’ll also be really easy to travel between. Going between these O’Neill colonies, from one to another—to visit friends, to visit family, to visit one that’s a recreational area—would require very, very low amounts of energy for quick transportation. It would be a day trip.

Professor O’Neill once appeared on television with the famous science-fiction author Isaac Asimov. The host asked Asimov a very good question: “Did anybody in science fiction ever predict [O’Neill colonies,] and if not, why not?” Asimov had a very good answer: “Nobody did, really, because we’ve all been planet chauvinists. We’ve all believed people should live on the surface of a planet, of a world. In my writing I’ve had colonies on the moon. So have a hundred other science fiction writers. The closest I came to a manufactured world in free space was to suggest that we go out to the asteroid belt and hollow out the asteroids and make ships out of them. It never occurred to me to bring the material from the asteroids in towards the earth where conditions are pleasanter and build the worlds there.”

Planetary chauvinists! Where will building this vision, these O’Neill colonies, take us? What will it mean for Earth? Earth will end up zoned residential and light industry. It’ll be a beautiful place to live. It’ll be a beautiful place to visit. It’ll be a beautiful place to go to college and to do some light industry. But heavy industry and all the polluting industry—all the things that are damaging our planet—will be done off Earth.

We would thereby preserve this unique gem of a planet, which is completely irreplaceable. There is no plan B. We need to save our Earth, and we shouldn’t give up on a future of dynamism and growth for our grandchildren’s grandchildren. We can have both.

Who is going to do this work? Not me. This is a big vision that will take a long time to realize. Kids in school today and their children will do it. They will build whole industries with thousands of future companies encompassing whole ecosystems. There will be entrepreneurial activity, unleashing creative people to come up with new ideas about how to use space. But those entrepreneurial companies cannot exist today. It’s impossible because the price of admission to do interesting things in space right now is just too high. Because there’s no infrastructure.

I started Amazon in 1994. All of the heavy-lifting infrastructure needed for Amazon to exist was already in place. We did not have to build a transportation system to deliver packages. It existed already. If we’d had to build that, we would have needed billions of dollars in capital. But it was there. It was called the US Postal Service, Deutsche Post, the Royal Mail, UPS, and FedEx. We got to stand on top of that infrastructure. The same was true of payment systems. Did we have to invent a payment system and roll that out? That would have taken billions of dollars and many decades. But no, it already existed. It was called the credit card. Did we have to invent computers? No, they were already in most homes, mostly to play games, but they were there. That infrastructure already existed. Did we have to build a telecom network, requiring billions more dollars? No, we didn’t. It was in place, mostly to make long-distance phone calls and built by global telecom carriers like AT&T and their equivalents around the world. Infrastructure lets entrepreneurs do amazing things.

Those O’Neill colonies will be built by today’s kids and their children and grandchildren. The job of building the infrastructure so these colonies can be created will start with my generation. We’re going to build a road to space, and then amazing things will happen. Then you’ll see entrepreneurial creativity. Then you’ll have space entrepreneurs start a company in their dorm room. That can’t happen today.

So how are we really going to build O’Neill colonies? Nobody knows. I don’t know. Future generations will figure out the details. But we do know that there are certain gates to pass through, certain prerequisites to meet. If we don’t do these, we will never get there, and it’s really nice to know what those things are because you can work on them, with great confidence that they’re going to be useful. However the details of that future vision evolve, two things will be essential. First, we must have a radical reduction in launch cost. Launches are simply too expensive today. And second, we have to use in-space resources. Earth has a very powerful gravitational field, and lifting all of our resources off of Earth just isn’t going to work. We need to be able to use resources that are already in space.

Named after Mercury astronaut Alan Shepard, the first American to go to space, New Shepard is Blue Origin’s reusable suborbital rocket system designed to take astronauts and research payloads past the Kármán line, the internationally recognized boundary of space. New Glenn, named for astronaut John Glenn, is a single-configuration, heavy-lift launch vehicle capable of carrying people and payloads routinely into Earth orbit and beyond.

One thing I’m very excited about with New Shepard is using it to get a lot of practice. The most-flown vehicles may fly a couple of dozen times a year, launching payloads into orbit. You never get really good when you do something just a couple of dozen times a year.

Let’s say you’re going to have some surgery. You should make sure the surgeon does that operation at least five times a week. Real data backs up the fact that a surgery is much safer if your surgeon is practicing it at least five times a week. And so we need to be going to space very frequently in a very routine way. One reason aviation is so safe today is because we do have so much practice.

We need to have more missions. If your payloads cost hundreds of millions of dollars, they actually cost more than the launch. This puts a lot of pressure on the launch vehicle not to change, to be very stable—reliability becomes much more important than cost. This actually drives you in the wrong direction of having fewer launches and very expensive satellites, and that’s what you see happening in many cases.

At Blue Origin we want to try to get on that practice groove, and to do that we have to have an operable, reusable vehicle. The key point here is operability. The space shuttle was only reusable in the most daunting of senses. In reality, NASA would bring the space shuttle back, inspect it in very elaborate ways, and then refly it. It would’ve been better to have an expendable vehicle. You can’t fly your 767 airplane to its destination and then x-ray the whole thing, disassemble it all, and expect to have acceptable costs. And so reusability vis-à-vis the airplane, versus the air shuttle, is really the key. Our goal is to drive down costs using reusability, and the vision is to figure out how there can really be dynamic entrepreneurialism in space.

I’m incredibly proud of the amazing progress the Blue Origin team has made on reusable launch vehicles with New Shepard. We have had eleven consecutive landings. We’ve used two boosters. One has flown five times consecutively, and the other has flown six times consecutively. There has been almost no refurbishment between flights. That’s how you reduce launch cost. You need to have reusable vehicles. Until now we have been using launch vehicles one time and throwing them away. You also can’t have fake reusability, where you bring the vehicle back and then do a lot of refurbishment, which is also very expensive. It’s incredibly exciting that we’re going to be flying humans on New Shepard soon.

When we built New Shepard, a suborbital vehicle designed for space tourism, we made some very curious technology decisions. It is, first of all, powered by liquid hydrogen, the highest-performing rocket fuel but also the most difficult to work with. It’s not needed for a suborbital mission, but we chose it because we knew we were going to need it for the next stage. We wanted to get practice with that hardest-to-use but highest-performing propellant. We made the same decision with vertical landing for New Shepard, even though other landing mechanisms would have worked at this scale. The great thing about vertical landing is it scales up really well. It’s very counterintuitive, but the bigger the vehicle, the easier vertical landing gets. Vertical landing is like balancing a broom on your fingertip. You can balance a broom, but try balancing a pencil. The moment of inertia of the pencil is too small. Right from the start, we wanted to build a human-rated system so we would be forced to think clearly about safety and reliability, escape systems—all the things we knew we would need to have practice with in order to build our next generation of vehicle. So it’s all about practice.

New Glenn is New Shepard’s big brother, big enough that New Shepard will fit in the payload bay of New Glenn. It’s a very large vehicle, with 3.9 million pounds of thrust. I get asked a very interesting question from time to time: “Jeff, what’s going to change over the next ten years?” And I enjoy playing with the answer. That’s a fun dinner conversation. But there’s an even more important question I almost never get asked: “What’s not going to change over the next ten years?” And that question is so important because you can build your plans around those things. So I know for a fact that Amazon customers are going to want low prices ten years from now. That’s not going to change. Customers are going to want fast delivery. They’re going to want big selection. So all the energy we’ve put into those things will continue to pay dividends. It is impossible to imagine a customer coming to me ten years from now and saying, “Jeff, I love Amazon. I just wish you delivered a little more slowly” or “I just wish the prices were a little higher.” That’s not going to happen. When you can figure out the things that are going to remain true under almost all circumstances, then you can put energy into them. We know what those things are for New Glenn. It’s cost, reliability, and on-time launches. Each one needs improvement before we can enter the next phase of really going out into the solar system, and I know these things are stable in time. We are not going to have a New Glenn customer come to us in ten years and say, “Jeff, I wish the rocket just, you know, failed a little more often” or “I wish it was more expensive or that you were late on my launch dates.” By the way, availability and launching on time are really big problems underappreciated by most people who aren’t directly in the space industry. Delays really snarl things up and cost the payload customers a lot of money. So these things won’t change. We’re going to put energy into them. The whole vehicle is designed around those three things.

Reusability is absolutely the key to radically reducing launch costs. People sometimes wonder how expensive the fuel is and whether the fuel is a problem. Liquefied natural gas is very inexpensive. Even though there are millions of pounds of propellant on New Glenn, the cost of fuel and oxidizer is less than $1 million—insignificant in the scheme of things. The need to throw hardware away is the reason why launching into orbit is so expensive today. It’s like driving your car to the mall and then throwing it away after one trip. That’s going to make trips to the mall very expensive.

The second gate we must go through is in-space resources. We have to use them, and we have a gift: this nearby body called the moon. We know a lot now about the moon that we didn’t know back in the Apollo days or even just twenty years ago. One of the most important things we know is that there’s water, an incredibly valuable resource, on the moon in the form of ice. It’s in the permanently shadowed craters on the poles of the moon. You can use electrolysis to break down water into hydrogen and oxygen, and you have propellants. Another great thing about the moon, another reason it’s a gift, is that it is nearby, just three days away. And you don’t have the same launch constraints, like the twenty-six months between launches that we have with Mars. You can go to the moon just about any time you want. And, of great importance for building large objects in space, the moon has six times less gravity than Earth. When you get resources from the moon, you can get them into free space at very low cost. It takes twenty-four times less energy to lift a pound off the moon than it does off the Earth. That is a huge lever.

But the moon also needs infrastructure. One way to build infrastructure will be through vehicles such as Blue Moon, a very large lander we have been working on for several years that soft-lands in a precise way, 3.6 metric tons onto the lunar surface. The stretched-tank variant of it will soft-land 6.5 metric tons onto the lunar surface. The deck is designed to be a very simple interface so that a great variety of payloads can be placed onto the top deck and secured. With its davits system, inspired by naval systems, things are lowered off the deck onto the surface of the moon. And the davits can be customized for the particular payloads.

There’s a lot of interesting science to be done on the moon, especially on the poles, and Blue Origin has formed a Science Advisory Board to make sure the science gets done right and we get the most bang for our buck. We also have customers for Blue Moon, and they are going to be deploying science missions to the moon as well. People are very excited about this capability to soft-land their cargo, rovers, and science experiments onto the surface of the moon in a precise way. There is no capability to do that today.

Vice President Mike Pence said that it’s the stated policy of the Donald J. Trump administration and the United States to return American astronauts to the moon within the next five years. I love this. It’s the right thing to do, and for those of you doing the arithmetic at home, that’s 2024. And we can help meet that timeline. It’s time to go back to the moon, this time to stay.

We must have a future of dynamism for our grandchildren and their grandchildren. We cannot let them fall prey to stasis and rationing, and it’s this generation’s job to build that road to space so that the future generations can unleash their creativity. When that is possible, when the infrastructure is in place for future space entrepreneurs, just as it was for me in 1994 to start Amazon, you will see amazing things happen, and it will happen fast. I guarantee it. People are so creative once they’re unleashed. If this generation builds the road to space, builds that infrastructure, we will get to see thousands of future entrepreneurs building a real space industry, and I want to inspire them. This vision sounds very big, and it is. None of this is easy. All of it is hard, but I want to inspire you. So think about this: big things start small.