Waymo is giving rides in Phoenix and nobody

The future arrived in Chandler, Arizona, and nobody noticed.

Waymo started offering driverless rides through its Waymo One service. Truly driverless. No safety driver behind the wheel. No human hand hovering over the emergency stop. Just you, a car, and an algorithm that’s driven more miles than you ever will.

The Arizona Republic covered it. A few tech blogs picked it up. And then… nothing. The news cycle moved on to whatever outrage was trending that afternoon.

I think we have a pattern recognition problem. We’re really good at noticing the future when it arrives with a bang. The Moon landing. The iPhone. The fall of the Berlin Wall. We’re terrible at noticing when it shows up in a minivan in suburban Arizona.

What actually happened

A person in Chandler, Arizona opened an app on their phone. A car pulled up. Nobody was inside. The doors opened. They got in. The car drove them where they needed to go, navigating intersections, stop signs, pedestrians, other cars. It dropped them off. It drove away.

That’s it. That’s the whole thing.

And it’s the most significant transportation development since the Model T.

I realize that sounds like I’m exaggerating. I don’t think I am. The Model T made personal transportation accessible. Self-driving cars make personal transportation… optional. You don’t drive. You don’t even need to know how. You just go.

The quiet arrival

I’ve been thinking about why this didn’t land as bigger news, and I have a theory.

Self-driving cars have been “coming soon” for so long that when they actually arrived, people had already processed the idea. It’s like how science fiction prepares us for technology. By the time the technology exists, we’ve already imagined it so many times that the real thing feels like an echo.

The thing is, there’s a big difference between imagining something and living in a world where it’s real. I can imagine a self-driving car. I’ve seen a hundred demos. But there are now people in Chandler who use self-driving cars to get groceries. It’s mundane to them. Tuesday to them.

That’s when you know a technology is real. Not when it’s on a stage. When it’s boring.

The Chandler, Arizona test

Here’s a detail I love. Chandler was chosen partly because of its wide, gridded streets and its dry, sunny weather. The conditions are about as perfect as you can get for a self-driving car. Not a lot of rain. Not a lot of snow. Not a lot of complicated intersections or old narrow roads or pedestrians jaywalking in unpredictable patterns.

Critics point this out. “Of course it works in Chandler. Try it in Boston.” And they’re right. Chandler is easy mode.

But every technology starts in easy mode. The first airplanes flew in clear weather over flat ground. The first computers solved problems that people could also solve by hand, just slower. You start where the conditions are favorable and you expand from there.

What matters isn’t that Waymo chose the easiest possible city. What matters is that even in the easiest possible city, with the most favorable possible conditions, a car with no driver just… drove. In public. With real passengers. On real roads with real traffic.

Five years ago, that wasn’t possible anywhere on Earth, under any conditions.

What I can’t stop thinking about

Here’s what keeps me up at night about this.

Waymo’s cars drive better than humans. The data bears this out. Fewer accidents per mile. Fewer near-misses. No road rage, no texting, no driving drunk, no falling asleep at the wheel. The machines are better at this.

So at some point, and I don’t know when, but at some point, someone is going to have to answer this question: if self-driving cars are safer than human drivers, is it ethical to let humans drive?

Not today. Not next year. But eventually.

And then the question after that: if we accept that machines can be trusted with our lives on the highway, what else can they be trusted with?

I don’t know the answers. I’m not even sure I’m asking the right questions.

The sci-fi comparison

In I, Robot (the Asimov book, not the Will Smith movie), there’s a story about the first self-driving car. The car works perfectly. The problem isn’t the car. The problem is that humans don’t trust it. They’d rather have a human driver, even knowing the human is statistically more dangerous, because at least they can yell at the human. You can’t yell at an algorithm.

We’re living in that story now. Waymo’s cars work. The people of Chandler who use them regularly report high satisfaction. The safety record is better than human drivers. But most people, if asked whether they’d get in a car with no driver, say no. Not because of the data. Because of the feeling.

I think that gap, between what the data says and what the gut says, is going to be the defining challenge of autonomous technology for the next decade. Not the engineering. The trust. The slow, grudging, deeply human process of accepting that a machine can do something better than us, and being okay with it.

We’re not there yet. We might not be there for a while. But something important just happened in a suburb of Phoenix, and I think we should be paying more attention.

The future doesn’t always announce itself. Sometimes it just pulls up and opens the door.