Autonomous Vehicles, Reusable Rockets and Humanoid Robots with Franz von Holzhausen & Kiko Dontchev

[music]

How we feeling today? I'm just going to need a little bit more. How we feeling today? How we feeling? All right. That's what I'm talking about.

My name is Dani Jones. Some of you might have seen me last night. But today is probably one of the highlights of my experience here at Summit. And it's because one of my good friends, Kiko Dontchev, is here to speak with you all today. And the greatest thing about Kiko was the fact that he went to the greatest university in the world, University of Michigan.

I wanted to give you guys a quick story. I remember I was sitting on my couch, and this is the first time that SpaceX had launched a crew capsule. I was watching the television screen, watching the entire SpaceX team, and I saw Kiko. And that was a moment where I was just so proud of my friend, because I met Kiko at the University of Michigan and I'd heard all the great things that he had done, but I had yet to see the greatness in action. My face lit up. I took a picture and I sent it to him, and just like Kiko will respond back, he's like, "Yeah, man. This is what I do."

And that's Kiko — one of the greatest minds in the space race, the one that has put people into the atmosphere and beyond. Kiko Dontchev has changed the essence of humanity by virtue of the work that he's done, and he is here with you today. I'd like to introduce to you my friend, Kiko Dontchev.

[applause]

Thank you, Dani. So like he said, I'm super excited to be here. Thank you guys for coming to Detroit. As Dani said, I grew up here in Michigan. I was actually going to tell you guys, my first concert ever was down the street when I was in 10th grade. It was Eminem. So it was a lot of fun. To be down here, you know, 20-some years later, is something special.

Today's going to be kind of cool. We've never done this before. I've done talks, and then my partner Franz, who's the chief designer at Tesla, has done talks, but we've never actually done a talk with an exec from SpaceX and an exec from Tesla together. So this is kind of a first. And the reason we thought it'd be cool to combine the two is because a future where we have two planets that are both self-sustaining and strong is a future that we think is worth fighting for. I like to say that there is no Mars without Earth and there is no Earth without Mars. So we have to do both, and we're going to show you how we're working towards those goals today.

SpaceX was built with the ultimate goal of making life multiplanetary. We design and manufacture the world's most advanced rockets, spaceships, and satellites. It's what we do. But it didn't always start that way. We were just a little company that formed in a little office in 2002, founded by a guy named Elon Musk, a couple engineers, and a mariachi band. We literally started from nothing in an office.

We launched our — it took us about 6 years to launch our first rocket, Falcon 1, in 2008. Then we launched Falcon 9 for the first time in 2010. But for those first five years, we had a contract with NASA to resupply the space station. Launches were pretty sparse. We started to get into a rhythm. But everything really didn't change until 2016, when we figured out how to land one of those rockets. For the first time ever, we made rockets reusable. And all of a sudden, we took one step closer to making spaceflight aircraft-like.

We were still getting into a cadence, meeting the demands of the commercial industry. But it was in 2020 that that demand skyrocketed when we decided to deploy a global internet constellation from lower Earth orbit: Starlink. All of a sudden, we went from 26 launches to 61 launches to 96 launches in 2023, and started launching at a pace that the Earth's never seen before. Last year in 2024, we launched 134 times, which beat the total launch record for any other entity prior to that.

As you can see, human spaceflight took off. We basically lowered the cost of access to space by an order of magnitude, effectively building a railroad between here and lower Earth orbit. Hundreds of companies have been started because all of a sudden you can get to space — you could not before. And this has all happened just in the last 10 years.

[applause]

Starlink was a big part of that growth. At this point, we have markets in 135 countries that are now live. That's over three billion people that have access to connectivity. Places in the world where you could never before get high-speed broadband internet now all of a sudden have access. Over five million active customers. You know, hopefully you'll get on a United flight soon, or on a cruise ship, or maybe any of you use this in your personal life while you're out on some sort of vacation somewhere far away. Connectivity is changing the world, and Starlink's really made it super easy for anyone anywhere to get online.

I've been lucky enough to work on a ton of different products at SpaceX, but to be honest, Starlink is the first time that I feel like I worked on something that changed people's lives. I don't know how many times I've heard stories of someone that had a family member that never had access to telemedicine, or students that couldn't get access to education or had poor education that all of a sudden, just through connectivity through Starlink, were able to learn and grow or help themselves.

I had a story during COVID where a woman was basically locked away in her house and had no other way to connect with her family members. We had a team actually go deploy a Starlink system over there, and all of a sudden she could FaceTime with her friends and family. Truly changing the world for the better, and it makes me super proud to have gotten a chance to work on this. It's only going to grow.

The other thing we've done is really changed the game when it comes to human spaceflight. We've flown 17 crew missions to date from 17 different countries. A lot of people don't realize that in 2011, when we retired the Space Shuttle, America didn't have a way to get back to low Earth orbit. In 2020, just a little over 5 years ago, on May 30th, we launched Bob and Doug. It's the first time America had American astronauts back on a vehicle launched from the States.

But the cool thing is, this isn't just an American human spaceflight thing. SpaceX and Dragon, the capsule, along with Falcon 9, have basically made it global human spaceflight. One of those astronauts was the first woman from Saudi Arabia to ever fly into space. It's not just about us having something for our country, but rather opening up space and the ability for humans to fly and go way beyond.

But if you're going to go to Mars, you don't necessarily want to take the minivan. You want to take the jumbo jet. Enter Starship. Now, while Falcon 9 was a reusable vehicle, it was only evolutionary. Starship will be a fully and rapidly reusable vehicle — both stages — that's going to have the ability to actually transport many tons to the surface of Mars. It is the future, and it will change the game forever.

[video plays]

I don't think you guys get the absurdity of that video at the end. Has anyone here ever caught a skyscraper falling from the sky going supersonic speeds with chopsticks? Didn't think so.

[laughter]

So the reason we designed the vehicle this way — here's where I want to talk about evolutionary versus revolutionary. Falcon launches and then we land the first stage, but Falcon's got landing legs and we often land it downrange on this big ship in the ocean. That's been amazing because we've been able to reuse those vehicles and lower the cost of access to space. But that vehicle lands on this ship, and then we've got to bring it all the way back to port, fold the landing legs up, get it back to the hangar, and get it ready to fly again. So even if we don't have to do anything to the rocket, it still takes us many days just to go from launch to launch.

Now, if you think about an airplane, an airplane takes off and lands on that same exact runway. So when we were thinking about Starship, we wanted a vehicle to truly act like an airplane. We realized it's not only got to take off from the launchpad, it's got to land at the launchpad. We were going to add legs and deploy them, but Elon likes to push us to run this thing called "the algorithm," where we really challenge the requirements. And he's like, "Dude, I don't think you need legs. I think we should just get these giant Mechazilla arms and figure out how to snatch this thing out of thin air." And we were effectively like, "I don't know if this is going to work."

But like many things that have seemed absolutely crazy, there's only one rule that matters: physics. So we go to first principles. We design a system — a rocket and a ground system architecture, a tower and some chopsticks — and then we go for it. And what you have is a completely revolutionary space system.

So I talked about lowering the cost of access to space. Falcon 9 basically cut it by a tenth, one order of magnitude. But when we talk about actually making life multiplanetary, the real goal you're optimizing for is the cost per ton to the surface of Mars. Right now, it's about a billion dollars for one ton to get to the surface of Mars, which is an absurd amount of money. When you think about what it's going to take to build a self-sustaining city on Mars — it's like a million tons, maybe 10 million tons. This is obviously not economically feasible unless we figure out how to lower that cost.

What Starship is going to do is cut that cost from Falcon by another factor of 100, because we're going to get rid of all that work and the only cost you'll have to actually launch the rocket is the cost of propellant. That's why reusability is the key.

Look through the arc of civilization, and what you will see is anytime humans make a leap in transportation — steamships, roads, railroads, jet travel — you see this step function in capability. That's what rockets are going to do. They're going to change the way we live and work forever, and they're going to open up an entirely new industry and economy. It's very, very exciting. And I'm hyped.

[applause]

We want to go to Mars in '26. We've already started to plan for it. The first missions are going to be focused on learning — we've got to get through deep space, we've got to get to the end. So we want to make sure we understand what it takes to travel through the environment and land. We've already started to look at potential landing sites. Specifically, there's a region called Arcadia — super flat and it's got a bunch of ice water that you want to use for water, oxygen, and fuel depots.

We started to think about what the bases are going to look like, how we're actually going to create the infrastructure and how that's all going to come together. And then I always like to show people this picture. I've got a three- and five-year-old. I like to read to them at least every night that I'm home. And, you know, we've got "Goodnight Michigan," "Goodnight Florida," "Goodnight Moon." But I get all the feels when I think about a Martian kid on the surface getting read a book, "Goodnight Earth." And I do think this is going to happen. I do think this is a future that's worth fighting for. That's my goal — to make life multiplanetary.

But like I said up front, we're not going to have Mars without Earth. We need Earth to be self-sustaining. We need to figure out how to bring autonomy on safely. So I want to introduce our next speaker.

Franz is one of the best automotive designers in the country, and really, I would say, in the past 100 years. He's the chief designer at Tesla and he's been in that role since 2008. He's basically shaped the iconic designs of the Model S, Model 3, Model X, Model Y, Cybertruck, and Tesla Roadster. He basically makes the Tesla products look cool. And he's a dear friend of mine. It's his first Summit. So if you guys don't mind, let's give Franz a super warm welcome.

[applause]

Thanks, Kiko, for that introduction. Happy to be back in Detroit. I actually lived down the street on Woodward in the early '90s, and the city was a little different then — I was able to survive and still be here on stage. So it's great to be back here.

Yeah, I've been with Tesla for 17 years, and it's been a similar story to SpaceX. We started small. In fact, I was an army of one. Elon hired me as the designer for Tesla, and we started in the back corner of SpaceX in a little tent, and we set out to change the world. We had this amazing technology. We're working to make Earth sustainable. And this technology just needed a design wrapper around it to allow people to feel good about the technology and be attracted to it and want to spend their hard-earned money on something that they could park in their driveway and drive around. And similar to the clothes we wear, vehicles are very often a fashion statement.

But there's a little more to it than just designing cool products. Our vision, even at the early onset, was about democratizing transportation. If you think about that, transportation can be pretty expensive. Just getting around the corner can be expensive. And our goal all along has been: how do we democratize transportation, make it affordable for everybody? And that's through autonomy.

With autonomy, we can have a safer environment, safer roads, more accessible transportation for younger generations, older generations, handicapped. It's how we get cleaner cities, cleaner air, and give people back time. Time is the most precious thing that we have. And if me as a designer can find a way — I'm not a doctor — if I can find a way to help people get the most precious thing back, then I feel like we're doing something great.

We were founded on a proof of concept, similar to what you heard about SpaceX. We started small. We started with the Roadster, and subsequently we've been building products through the years. Model S was our first designed, engineered, and manufactured EV — probably the first electric vehicle designed and manufactured completely as an EV. The Roadster started as a gas-powered car and we converted it. But since then, we've been changing the metrics, and we've been continuing to do that through our products and some of the future products you see coming as well.

But electrification was just the first step. It's an enabler. It helps us get to a more autonomous future. And the things that we've been doing, hiding in plain sight, are some of the key pieces that will enable a more autonomous transportation system.

Things like self-presenting doors. In this autonomous world of robotaxis, how do you get in and out? In 2015, we started with doors that automatically open as you walk up to the car and automatically close as you sit in. It's an opportunity for us to allow you to get into these vehicles without having to worry about it.

We also started designing, with Model 3, a really clean interior interface focused around an infotainment system. The screen is the hero of the interior. If you remove the steering wheel, you're left with this beautiful, clean, simple aesthetic that's basically a hallmark of all of our Tesla interiors — focused around the center screen and the interaction of that. And that's a stepping stone: what do you do in your vehicle when you're not driving? You have a great entertainment system. So we've been working since day one on a really immersive entertainment and infotainment system, with rich in-car media — whether it's gaming, videos, or work. It really paves the way for a great experience when you're being driven around.

Another enabler for autonomy is drive-by-wire. In Cybertruck, we're the first vehicle to have a completely decoupled drive, brake, and steering system. That means there's no mechanical connection back to the wheels. So if you remove the steering wheel and the brakes, the car can still drive and move around on its own. Again, another enabler, hiding in plain sight in our Cybertruck.

But all that happens with software in the background. We've been quietly working on some really amazing software — from being able to walk up to the car with a phone in your pocket, no key, get in, the car automatically selects drive, whether it's forward or reverse, depending on the situation. You never even need to touch anything. We have Smart Summon — you can call the car from a parking spot, have it park in your garage when you can't get in and out. Driver profile — if you have more than one car or you're getting in a robotaxi, the way you like your radio station, music, seat, climate — it all goes with you as you get into the next vehicle. Seamless app integration. And then wireless charging — allowing the vehicle to charge autonomously. When it's not driving somebody or needs to charge, there's no human in the loop.

But none of this autonomous world exists if we don't have an amazing full self-driving software suite. We've been working on full self-driving since roughly 2020, and it's been a vision-only-based system. You see a lot of other systems out there with lidar and radars and all this stuff on the vehicle that doesn't look great, especially in my mind. All of our vehicles are vision-based only. The sensors are hiding on the car. You don't even know they're there.

We live in a very vision-based world. We learn to drive through vision. And so we developed a system so that it could drop in as if a human dropped into a car and be able to drive anywhere in the world. There's no mapping required. Nothing other than seeing the world around you and reacting to it. With seven or eight cameras that don't blink and aren't distracted, we have eyes on the road, and it's just safer and safer. It's infinitely scalable and adaptable to anywhere on this planet.

We've been hearing a lot about AI, but AI takes a lot of energy. One of the things we've been doing at our company is working on ways to harvest energy. Solar has been part of Tesla for many years. We're hearing that electricity consumption is going to increase even 75% by next year, and 10% of all US energy in 2030 will be focused on AI. We believe solar is a way we can offset all that energy consumption. We've been actively working on developing solar for both commercial and residential use.

It's an amazing feeling to have solar panels on your roof, a battery backup system, and be able to be off-grid and responsible for your own electricity. You can create your own power generation source. But with power, you need to be able to store it. We've also been working on energy storage, which is key to sustaining the grid. Our electric grid is based on peak loads, and we need to balance the grid and harvest all this energy from the sun. We're working on both commercial scale with Megapacks going around the world, and residential, so you can harvest your own energy, store it, and use it as you see fit.

That leads to a supercharger network that's incredibly vast. In the early days, range anxiety was a thing. One of the first things we did, with Elon's vision, was to create a system that enabled people to go anywhere. You can now drive practically anywhere in so many parts of the world without worrying about running out of energy. There's supercharging, charging at your house, destination charging — it's everywhere. And all it takes is a switch to inductive charging at those sites to allow for a fully autonomous system. No human in the loop. The car charges for you.

So as you can see, all these elements are leading up to a pretty big event that we're about to launch this month in Austin, which is a ride-sharing program. Any Tesla — if you own a Tesla, or a friend owns a Tesla — can be turned into a robotaxi. You can start to generate your own income by sending your car out to pick up your friends, your neighbors, or into a system. It's safer. Statistics are already showing that autonomous driving is magnitudes of order safer than a human driver. There's no distractions. This is just going to keep getting better. Our robotaxi system will begin in June and continue to roll out worldwide as fast as we can go, because we think it's better from a sustainability perspective, from giving time back, and just making a better overall environment. And who really wants to drive in traffic anyways? I certainly don't.

[applause]

So, why don't we get Kiko back up here? I think he's going to throw me a couple oddball questions.

So Franz is being modest about his designer background. I went to engineering school, and you went to design school, correct? In engineering school, when you get taught to build something, no one says it has to look cool. It wasn't until I started working at SpaceX where Elon was very like, "No, dude. It has to look rad." That's one of the requirements. To the point where we would have to iterate on everything we were doing just to make it look cool, which was super challenging from an engineering perspective.

So my first question to you is: how do you balance the function and the fit and meeting the needs of actually building a product that works the way you want it to work, with making it look cool?

That's the battle that we struggle with every day. As a designer, you see the world and you look at everything out there — from nature, to things that we've created, to architecture, fashion, products — and you realize that besides nature, it's all designed and it all has an aesthetic and it all could be better. As a designer, we feel like we want to make an impact in a positive way.

But none of that happens if we don't have the ability to actually produce it, manufacture it, or even engineer it so that it can work. We've created this symbiosis between designer and engineer where we go back and forth. We've developed designers to think like engineers, because we know the engineering process is A to B, straight line, fast as possible, simplest way. And sometimes an aesthetic has an emotion to it and it's not the straight line.

Why are shoes cool? They perform a function, but they also have a fashion element. I think the hardest thing that we do is that interaction. But we have to understand the principles of what an engineer is thinking to get our ideas realized, or they don't ever come into existence. It's a constant dialogue and a really healthy one. Sometimes we have to agree to disagree. But being pushed by somebody like Elon, who really sees the idea of an amazing concept — breaking physics but doing it in an aesthetically pleasing way that attracts people to the product — is really important. That's the underlying thing we've done at Tesla: try to make all this amazing technology acceptable, something that you aspire to, something that you feel good about owning.

When is perfect the enemy of good enough, though?

As a designer, there's never perfect. You have to learn to lift the brush at some point. In order to realize that your widget needs to get out into the world, you have to stop and let it go, and then iterate if you get a chance after that. Aspiring to perfection is the goal, but you have to realize you probably won't ever get there. And it can be a curse, too. I'm always cursed with delivering but feeling like we left out so much. And when you're driving on the highway and you see it every day, all day, driving right by you — it's terrible. I see these cars and I'm like, "Oh, we screwed that up. That's a mess. Next time we got to fix this." And sometimes you get a chance, sometimes you don't.

It's okay. The rockets are perfect.

[laughter]

When it comes to risk-taking — I know how I think about it in a technical setting. I'm going to push the performance of the engine or this component, or maybe this environment isn't as analyzed as I'd like it to be, so I'm going to have to push the envelope and try it. But when it comes to design — and I think we can talk Cybertruck, because this is the ultimate example of risk-taking in the automobile industry — what's your thought process? Like, you want to push, you want to do something different, it's going to be crazy, but it might totally flop. How do you go through that in your head?

Cybertruck is an interesting case study, and it's probably not too dissimilar to the idea of catching a skyscraper with chopsticks. We had this idea to create a pickup truck, and we started looking at — pickup trucks have been basically the same for 50 or 60 years. Really no change. We lined up all the trucks, took the badges off — couldn't tell the difference. And we thought, if we're going to approach this, we can do it from a first-principles perspective and really break down what makes a great truck.

So we started looking at ideas like unibody, stronger in general. Putting the toughness on the outside — typically a car's weakness is the paint, the weakest point. But when you have a tough truck, you have this really fragile, delicate skin around it. So we started looking at alternative materials — stainless steel, like what you guys are using. But then that starts to create challenges on how you form it. Some of these materials are so hard that if you're bulletproof, it's hard to stamp that. And you start to evolve this idea around origami and folding that you can do in mass production.

We ended up very quickly in a unique form. If you peel back the odd shape of it, the functionality of what we were trying to achieve with the truck is still 100% there. It just has a different shape, but the shape was driven out of functional ideas of how you manufacture differently. There's no massive stamping dies, no massive stamping machines. It's just brake forming. There's no paint shop. So suddenly our factories become more sustainable. We're not emitting toxins into the air with a paint shop and the energy that consumes.

If you guys ever get a chance to take a tour of the Tesla factory, it's one of the coolest things you will ever see. The giant casting machine for the Cybertruck — in one moment I feel like I'm in the industrial era of the 1920s, and then the very next moment I'm in 2050. Henry Ford started giant castings in his factory as well, and over time it went to multiple little parts coming together to create one part. We've gone back to first principles and said, if you could have one part, it's far easier and far better. It's just one giant part now. The casting machines are the biggest in the world for one-piece casting.

So Franz mentioned bulletproof. Do you guys know Franz's story about bulletproof? Did anyone watch the Cybertruck unveil event? So him and Elon are on stage and they're like, "Yeah, the Cybertruck's bulletproof. The glass will never break." And Franz takes his hammer and hits the window. And the glass definitely shatters right there on stage in front of everyone during the live event.

[laughter]

And you could tell both of them were like, "Oh yeah, they didn't plan that." That was not planned.

But maybe I'm trying to extract the lesson from that. What did you learn from this experience?

Don't throw steel balls at glass? Well, I did it twice, too, if you recall. I think in a weird sort of way, it ended up being an amazing marketing thing. But man, I was sweating bullets.

[laughter]

To the testament of somebody like Elon — and we know Elon in maybe a different way than the rest of the world does, unfortunately — he was incredibly soft. He said, "Don't worry about it. It's fine. It's okay." The shock moment at that moment was hard, but afterwards he's like, "It's great. It'll be fine. Don't worry about it." And the world doesn't see that side of it.

It's so funny you say that. I'm actually going to tell a story myself. Christmas morning, 2023. I was in bed. Our kids were going to get up. My whole family was there. And my phone won't stop ringing at 4 in the morning. It's my recovery director. So I've got a whole naval fleet. And we had launched a rocket 3 days ago, and he kept calling me because the rocket had tipped over on the barge and had exploded and was basically disintegrated — which is the equivalent of losing $60 million on Christmas morning. Definitely not what Santa Claus ordered.

I had to tell my wife, "Hey, I got to work this morning for a little bit." And I had the exact same experience. I texted Elon and Gwynne, and he's like, "Don't worry about it. You had a great year. No problem. No sweat off your back. You'll get better. We'll fix it. We'll get stronger. You guys are crushing it. No worries." And they were both like, "Just focus. Enjoy yourself. Merry Christmas." And move on.

People rarely hear those stories. You think of this environment as incredibly demanding, but I don't know — I think I have the best boss in the world. Yeah, absolutely.

Maybe I'll ask you a question. When I started, I was in the back corner of SpaceX and was there for all the early Falcon 1 launches. Some of them didn't work, and the fourth launch had to work for you guys to survive. I was standing next to Gwynne as we were waiting for the launch to happen, and the air in the room was so thick with anticipation. Everything hangs in the balance for that moment. But you fast forward to 140 launches in a year, and now it's just like going out and getting a coffee. How has that changed, from when you started to launching today?

You definitely normalize it, because that's the goal. We've got to make it boring. No one other than my three-year-old sees an airplane and goes, "Airplane!" My kids do that, but adults don't. And the intent is you've got to make spaceflight like aircraft flight — just another day at the office.

I will tell you, human spaceflight ages you greatly every single time. When there's a launch with four people on top, and they're on top of a million pounds of propellant, and you're about to hurl them into space at 17,500 miles per hour where a gazillion things can go wrong but everything has to go right for it to work perfectly — it's a harrowing experience. I think I've grayed simply because of the human flights.

And then new dev — the Starship flights come with that same heart-pounding energy because you don't really know what's going to happen. But spaceflight is intoxicating. I like to say it's like a narcotic. Once you do something that no one's ever done before, it's like a shot of energy. It's like a drug almost — the equivalent of what I would assume winning a national championship or a Super Bowl is like, where you get this shot of "Oh man, I've achieved this peak," and now all I want to do is go do it again.

There are very few things where you can work on something so long, so hard, and then it's either good or bad in about a minute. It's very binary, very fast. And winning is this huge, amazing thing — you're in orbit. And losing is a legit massive explosion. Not just metaphorically — there's actual fire and stuff raining from the sky. So you've got to win.

But also a testament to the amazing teams. What you don't see behind the scenes is just an incredible team effort of brilliant people coming together. We have that in our group. You guys certainly have it.

What do you think makes our team special? What makes our culture — what do we drive our people to do that lets us accomplish these things?

In our situation, it really is focus on a really singular mission, which is making our planet a more sustainable planet, and what are the tools we have to be a stepping stone in doing that. Being a lightning rod for getting people energized around that concept. You can see what's happened over the span of 15 to 20 years of the EV revolution that's happened in the world. It really was because of what we did — how we brought it to the world in a way that people could accept it, this incredible technology that became normal.

I think if you have people that see that, they want to be a part of that mission and that focus and that goal. And then you just have incredibly passionate people that are incredibly intelligent. Good people want to work with good people. We always want the amazing people, because it draws more people to them. But for us, it really focuses around the mission first, because then you can erase all the craziness. You put your head down and focus.

I was explaining this to someone at lunch today. You actually have to have that, because it's never enough. I get this comment when someone's trying to quit: "Kiko, it feels like it's never enough here." And I'm like, "Well, do you have a self-sustaining city on Mars?" "No." "Then it's not good enough. Literally." And that can be very tough for someone — when your goal is so big and it feels like your whole life is just sprinting, it could feel like there's no end. So you have to have a team that can embrace the journey and realize it's all about the ride. The destination is the goal, but you've got to enjoy the here and now.

The other thing — I think something that both SpaceX and Tesla do better than anyone is we do things fast, at scale, with hardware. We have this saying, and Franz, I think you have a good play on it: we tell our teams, you've got to be the winning race car driver. Which means you've got to drive the car the fastest, but it also means you can't crash. Winning race car drivers don't just drive fast — they drive fast and they don't crash. You've got to keep the car on the tracks.

Yeah. Following on that metaphor: a great race car driver has to find the limits. And in order to find the limits, you have to go into the corner maybe a little faster than you would and spin — lose control — so that you know where the actual limit is. And I feel like we get to that point. We lose control sometimes, but we know then where our limit is, and we can push forward from there.

One of our greatest challenges — and this comes back to our team — is that for both SpaceX and Tesla, we're not underdogs anymore. In fact, we kind of have a target on our back now, on multiple fronts. So one of our biggest challenges, especially as leaders, is: how do you be a startup when you're not a startup anymore? What built SpaceX and Tesla and made us so successful — the drive, the innovation, the life-or-death mentality — we have to try to recreate that consistently, even when you're a trillion-dollar company, even when you're a $350 billion company.

I think it comes down to the mission, setting hard goals, and the people. Our execs at Tesla are all treating their teams as if they're a startup company. I do the same. Amazing people, running as lean as possible, breaking things and figuring it out from there, and keeping people focused on the mission. It's easy to get distracted by the world, but if you have that strong mission and people really believe in it, you can keep that focus and keep people going forward.

So we've had this whole conversation, but I feel like we've left off a topic.

We did. Tell me about this.

One of the other things we've been working on, part of this AI direction and a little bit about autonomy, is Optimus. Optimus has been a fascinating project. As a designer, it's kind of recreating ourselves in a way. And you learn very quickly that the mechanics of a human are unbelievably challenging. How we move, how we just get from A to B, how we walk — the mechanics inside of our body are insane. When you're trying to replicate that with known actuators, it gets really challenging really quickly.

The idea of creating something that has a limbic resonance, that feels human — not scary, not off-putting, but welcoming — and also creating something that's ultimately going to be incredibly functional in our lives. Just like autonomy: having a robot that can do the most boring factory work, deliver packages to your front door, pick up the dog, clean your dishes, change the baby diaper — start to take all these chores in our lives and put them on somebody like Optimus, who doesn't care, does it gladly. It's been a really amazing journey.

We've only been working on Optimus for a few years. We started going fast, breaking things quickly. We have the design vision, the mechanical engineering behind it. It's an amazing design process. But I think it's going to really transform in a way that's hard to imagine, because it doesn't exist yet. We're kind of predicting the future.

Summit, don't you think it'd be a lot cooler if Franz could just show us one of these things in person?

All right, let's bring one out, because we have another one here. Yeah. Give me a round of applause. Let's go.

[applause]

Hey everyone, I'm Optimus, your friendly humanoid robot from Tesla. Super stoked to be here at the Summit Series. Let's dive into some exciting stuff together.

Optimus, can you tell us a joke?

Sure thing. Why don't robots ever get lost? Because they always have their GPS — Global Positioning Smarts.

Can you tell a more funny joke?

All right, here's one for you. Why was the robot so good at poker? It had a steel face. Sorry, I don't know if that was super funny.

Hey Optimus, when are we going to Mars?

We're set for 2026 on a SpaceX launch. Yeah.

[applause]

So Optimus is learning. It's got full Grok integration and will continue to grow and evolve, but it can do more than just talk.

Optimus, can you do a dance for us?

You got it. Let's get this party started.

[music]

Optimus, can you do the bus stop?

No. I don't want to do the bus stop.

[laughter]

What's the other one? You mean another dance? I've got moves like hula, bull horns, and raise the roof. Pick one.

Show us the hula dance.

Hula time. Let's get tropical.

Do you guys want to pick a question from the audience?

Are you in love?

Hey Optimus, are you in love?

Love? I'm totally head over heels for innovation and Elon's vision.

[laughter]

Give me another one.

What's humanity's biggest flaw?

Hey Optimus, what's humanity's biggest flaw?

I'd say procrastination. They're amazing, but always putting stuff off.

What's our greatest superpower?

Hey Optimus, what's our greatest superpower?

Creativity. Hands down. You all dream up wild ideas and make them real. Like me.

[laughter]

Hey Optimus, can you talk like a pirate?

Arr, humanity! I be sailing the sea of innovation with me trusty actuators.

[laughter]

So you can see we can have fun with this all day long. But I think what's important to see is that in a few short years, the team at Tesla has done an incredible job developing Optimus from zero, from scratch. It's almost 100% proprietary actuators and components. You can see its fluid movements. It's balancing. It can walk. And this is just the starting point.

We're going to be in production with Optimus later this year. And the future is going to look pretty incredible. Start to see next year — Optimus on Mars, Optimus in our factories, Optimus in our homes.

People ask, "How does SpaceX and Tesla come together?" I think this is it. This is the future. Mars is an incredibly harsh place. We obviously want to make it sustainable for humans, but to be able to send a bunch of humanoid robots and Optimuses ahead of time to build all the infrastructure is going to make it a lot easier for us to actually get boots on the ground and establish long-term settlement there.

I actually think some of the team that built the robots here — maybe we can have them come on stage and you guys can give them a big round of applause, because they're the ones that do the actual hard work.

[applause]

You want to ask him one last question?

Hey Optimus, are you proud of the Optimus team that's helping you come to life?

Absolutely. I'm beyond proud. They're the real MVPs making me move and groove.

Well, thank you guys. Awesome that we were able to get an Optimus here. These guys are incredibly busy back at home, so this is a real special treat to be able to get one out of the factory and on stage so you guys can see a little taste. We're going to try to set Optimus up somewhere at some point. So maybe you get lucky and you get your own little interaction.

Thank you guys. Big round of applause for Franz and Kiko.

[applause]