The History and Invention of Waymo

Listen in Sam's voice (generated with ElevenLabs)

Hey kids — today's story is about a car that drives itself. No hands on the wheel. No feet on the pedals. Sometimes, no steering wheel at all! It started as a crazy science-fair idea in the middle of the desert, turned into one of Google's strangest secret projects, survived a betrayal that ended up in court, and finally became a real company called Waymo that you can actually call with an app, like calling a taxi — except nobody's driving. Buckle up. This is the story of the self-driving car.

It starts in the Mojave Desert in 2004. The United States military had a problem: they wanted robots that could drive supply trucks through dangerous areas without risking a human driver. So they created a contest called the DARPA Grand Challenge and offered a million dollars to any team that could build a car to drive itself across 150 miles of desert, completely on its own. Fifteen teams showed up. Here's how it went: every single car failed. One flipped over at the starting line. Most got stuck, crashed into fences, or just stopped and gave up. The furthest any car got was about seven miles. Out of 150.

So the next year, 2005, they tried again — and this time a team from Stanford University, led by a professor named Sebastian Thrun, built a car named Stanley. Stanley was a modified Volkswagen stuffed with computers, cameras, and a spinning laser sensor on the roof. And Stanley didn't just do a little better. Stanley finished the entire course — all 132 miles, through desert and mountain passes, in under seven hours, with nobody inside. Stanley won the two-million-dollar prize, and more importantly, Stanley proved it could actually be done.

A few people at Google were watching very closely, including the co-founder, Larry Page. In 2009, Google quietly hired Sebastian Thrun and his best engineers from that Stanford team and gave them a secret project inside Google's most experimental division, nicknamed Google X — the 'moonshot factory,' where they worked on ideas that sounded like science fiction. The project's codename was simple: Chauffeur. Just like Stanley, these early cars used a spinning roof sensor called LIDAR, which is like a bat's sonar but made of laser light instead of sound — it fires out millions of laser pulses a second and builds a 3D map of everything around the car, down to the inch.

For years, almost nobody outside Google knew this project existed. Modified Toyota Priuses and Lexus SUVs, topped with that funny spinning sensor, quietly drove hundreds of thousands of miles around California — merging on freeways, stopping at lights, even handling a kid on a bike who darted into the road. A trained engineer always sat in the driver's seat, ready to grab the wheel if anything went wrong. By 2012, these cars had safely driven over 300,000 miles. The secret finally went public, and the world realized Google wasn't just a search engine anymore.

In 2014, the team did something bold: they built a car with no steering wheel and no pedals at all — just two seats, big friendly headlight 'eyes,' and a single button that said GO. People nicknamed it the Firefly. It could only go 25 miles an hour, and it looked more like a golf cart than a car, but it was the clearest possible statement: this wasn't about helping humans drive better. It was about replacing the need for a human driver completely.

But then came a betrayal. One of the project's star engineers, a man named Anthony Levandowski, left Google in 2016 to go start his own self-driving truck company — which Uber bought almost immediately. Google's team discovered that right before he quit, Levandowski had secretly downloaded over 14,000 confidential files about their LIDAR technology. Google sued Uber. It turned into one of the biggest trade-secret lawsuits in Silicon Valley history, and in 2018 Uber settled by handing over 245 million dollars worth of its own stock. Here's the lesson, kids: taking something that isn't yours doesn't just get you in trouble — it can cost you almost a quarter of a billion dollars.

Right in the middle of all that, in December 2016, Google spun the whole project out into its own independent company, under Google's parent company Alphabet. They gave it a brand-new name: Waymo — short for 'a new way forward in mobility.' No longer just a weird lab experiment. Now it was a real business with one giant goal: build the world's most trusted driver.

The biggest milestone came in October 2017, in a suburb of Phoenix, Arizona called Chandler. For the very first time ever, a Waymo car drove on a public road with an actual empty driver's seat — no safety engineer ready to grab the wheel, nobody at all, just real neighborhood families riding in the back seat of a minivan that was truly, completely driving itself. After eight years of secret testing, that was the moment the dream became real.

Then, in December 2018, Waymo launched something called Waymo One — the first driverless taxi service regular people could actually use. You'd open an app, just like calling a ride, and a car with nobody in the driver's seat would pull up and take you where you needed to go. It started small, just in part of Phoenix. But it kept growing — San Francisco, Los Angeles, Austin — and today Waymo cars have given millions of completely driverless rides and driven tens of millions of miles with no human controlling the wheel at all.

So why does any of this matter? Because most car crashes — the vast majority — happen because a human driver got distracted, tired, or made a mistake. A computer doesn't check its phone, doesn't get drowsy, and doesn't get angry at the car in front of it. If self-driving cars really can be safer than people, this could end up saving more lives than almost any invention in the history of transportation. But notice something important about this whole story: it took nearly twenty years, from a desert full of crashed robots in 2004 to cars quietly picking people up today. Nothing about this was overnight. It was two decades of failing, rebuilding, getting betrayed, and trying again.

Here's what I want you to remember. The team that won in the desert wasn't the one with the flashiest idea — it was the one that kept testing, kept failing safely, and never stopped improving, year after year after year. That's how a real founder thinks: patience is the whole game. Next time you're in the car with me, look around — maybe someday soon, nobody will need to be driving it at all. See you next time on the Lessin Boy Podcast.