The safest car on the road is the one with nobody inside. You can't injure a passenger if there is no passenger.
Self-driving cars are not about technology. They are about saving the 40,000 lives lost on American roads every year. Everything else is a footnote.
We started this project at Google in 2009 because we believed autonomous driving could be solved. Fifteen years later, we are the only ones actually doing it at scale.
The DARPA Grand Challenge proved autonomous driving was possible. Google proved it could work in cities. Waymo is proving it can be a business.
We are going to put a driverless car on every road in America. That is not a goal — that is an inevitability.
GM gave us $10 billion and told us to solve autonomous driving. We got close. Close is not enough.
I helped build the technology at Google that became Waymo. Then I left to bet on trucks, because highways are simpler and trucking is an $800 billion market.
We merged with Uber ATG because combining the best teams gives you the best chance. Self-driving is too hard for any one company to solve alone.
I have been working on self-driving cars since 2004. This is not a fad for me. This is my life's work.
The Origin was designed to be the first purpose-built autonomous vehicle — no steering wheel, no pedals, no compromise. The future just came a little later than we planned.
Autonomous vehicles need maps that update in real-time. That's a very different problem than showing you a pizza place.
We had the money. We had the talent. We had two of the biggest automakers in the world. What we didn't have was enough time.
The technology is not the hard part. The hard part is making the technology work in every situation a human driver would encounter.
Profitable, fully autonomous vehicles at scale are still a long way off. We're being honest about that.
The trucking industry is short 80,000 drivers. Autonomous trucks aren't replacing drivers. They're filling jobs nobody wants.
Highways are more predictable than city streets. That's why autonomous trucking will be solved before robotaxis.
An autonomous truck can drive 20 hours a day. A human driver is legally limited to 11. The math is not complicated.
We're not building a demo. We're hauling real freight for real customers on real highways in Texas.
The defense contract isn't a side project. Military applications push our technology harder and faster than commercial ever could.
Everyone wants to solve the hardest problem in autonomy. We chose to solve the most profitable one.
Walmart doesn't do pilots for fun. When they put autonomous trucks on daily routes, it means the technology works.
We removed the safety driver in 2022. Fully autonomous trucks on public roads delivering for Walmart. No one in the cab.
Fixed routes are our unfair advantage. Our trucks drive the same roads every day. The more they drive, the better they get.
We spent 15 years developing the technology before anyone cared. Then suddenly everyone cared all at once.
Tesla dropped us in 2016. They said they'd build it themselves. We're in 170 million cars. They're in about 5 million.
Self-driving is a marathon, not a sprint. We've been running since 1999. Most of our competitors started in 2016.
LiDAR is the only sensor that gives you a true 3D understanding of the world. Cameras can see. LiDAR can measure. That difference saves lives.
The question isn't whether autonomous driving will happen. It's whether it'll happen fast enough to justify the billions being spent right now.