Autonomous vehicles hold a lot of promise, removing one of the biggest daily time sucks for millions of Americans and reducing human error in driving. But even as manufacturers rush to perfect the software, recent headlines about crashes involving self-driving vehicles may give potential riders, and certain lawmakers, pause.
A March poll from the American Automobile Association found that 68% of Americas fear riding in an autonomous vehicle, up from 55% in 2022.
Still, Kyle Vogt, the CEO of General Motors's driverless car maker Cruise, aims to have a fleet of at least 1 million robo-taxis on U.S. roads by 2030 and in markets like Japan and Dubai. He says their wide adoption is inevitable because of how much time Americans will save, the environmental and safety benefits, and changing attitudes toward car ownership. "One of the greatest shifts that will occur in our lifetimes is going from driving to being driven," he tells Fortune. (Besides the robo-taxis, GM and Cruise plan to eventually sell self-driving cars to individuals.)
Last year, Cruise, which competes with Alphabet's Waymo, Amazon's Zoox, and Tesla, among others, started offering robo-taxi rides in San Francisco and soon added Austin and Phoenix. This year, Cruise will introduce the Origin, an autonomous car without pedals or a steering wheel, rather than the retrofitted cars it has so far used. "Within 10 years, driving is going to be a hobby like riding horses is today," predicts Vogt.
Fortune: How far are we from seeing self-driving cars taking people to and fro?
There's no point in driving yourself today, even though we all do. As a society, we're killing an entire football stadium of Americans yearly because of distracted driving. We've turned a blind eye to that for over a century because there wasn't a better alternative. Now there is one. I think the vast majority of people will get around major cities in autonomous vehicles instead of driving in five years. Within 10 years, driving will be a hobby like riding horses is today.
Do you worry that headlines about crashes involving self-driving cars, like Tesla's, could dampen the public's enthusiasm for them?
If people are lumping in driver-assistance systems (like Tesla's) with driverless cars, which are different, then we run the risk of that because of those headlines. What we see is that a lot of people are initially pretty nervous about getting into driverless cars. But within two or three minutes of riding around, they get it. Over the next year or two, I think everybody will know someone with firsthand, on-the-ground anecdotes that will change perceptions.
What needs to happen to spur the wider adoption of AV tech? More involvement from the U.S. government?
That industry is still nascent overall. So our biggest competitive obsession is not the other people building AVs but people taking human-driven rides. That's the behavior we've got to change. I'm deeply concerned that the U.S. is moving too slowly and that Chinese competitors could get a leg up from how their government is not just opening the doors from a regulatory and infrastructure standpoint but also investing heavily.
I can see the safety and fossil-fuel arguments for the rise of self-driving cars, but what about congestion? Doesn't this tech just replace one kind of car with another on the roads?
A lot of the congestion problems are because of how humans drive. Think about how cars accelerate in an accordion fashion at a green light or how people slow down during an accident on a highway and do the rubbernecking thing. Humans are so bad at driving.
You're starting with a fleet of robo-taxis. But won't people still want to own cars in the future, self-driving or not?
Owning your car will be seen as the ultimate luxury in the future, and some people will want that. But the reality is for the vast majority of Americans, it's going to be far cheaper to just push a button on your phone and have a robo-taxi come and take you where you want to go all the time, compared to owning your car and having to deal with parking, having garage space, taking it for maintenance, and keeping it clean.
What needs to happen to spur the wider adoption of AV tech? More involvement from the U.S. government?
That industry is still nascent overall. So our biggest competitive obsession is not other people building AVs but human-driven rides. That's the behavior we've got to change. I'm deeply concerned that the U.S. is moving too slowly and that Chinese competitors could get a leg up from how their government is not just opening the doors from a regulatory and infrastructure standpoint but also investing heavily.
Your test rides have been in snow-free cities like San Francisco, Austin, and Phoenix. What about markets where there is a real winter?
We're first trying to scale this up as quickly as possible, and we're starting in the warmer, Southern half of the U.S. mainly for that reason. We don't need to solve the worst-case blizzard to start adding real value to people's lives today. And we are working on a next-generation vehicle with the necessary technology to handle bad winter weather. Some of the sensing technology we put on these vehicles can already see through snow, fog, and rain.
Who are the early adopters of AVs so far?
Younger people, especially college kids, love this. We're doing a partnership with the University of San Francisco, where a lot of students are out and about at night, coming in from classes or going to dorms and things like that. Maybe there's some overlap there because the younger generations are less interested in the hassle of car ownership and are looking for an alternative to Uber and Lyft.
Are densely populated cities your sweet spot? Or could AVs be popular with someone who lives in downtown Dallas but works in Frisco, Texas?
They could. People really like cars and driving but I haven't met a person who says they like sitting in bumper-to-bumper traffic going down the highway. Driving is one of those things where until you see the alternative, it doesn't click for you. In the future, the car will work for you. So you sit in the backseat, watch movies you like, do work on the phone, and consume an alcoholic beverage if you want.
This isn't your first time at the entrepreneurship-inventor rodeo. You are one of the cofounders of the livestreaming service Twitch. What's different about your experience leading Cruise?
Probably 80% of the acts that a founder or CEO of a company does are the same from one business to another. I learned a lot at Twitch about how to be a better leader, manage through a recession, and recruit and retain good people.
You returned to Cruise's corner office in early 2022 after a few years as CTO, replacing the man who had been your successor. What's different about your second tour of duty?
Going from a company of zero to a few hundred people is very different than going from maybe 1,500 to a few thousand. In the early stage, I was very hands-on, writing code and out there in cars, debugging things with the engineers, and dealing with everything that needed to happen in the office. I used to take out the trash. My time now is much more about recruiting really good leaders. Managing an experienced C-level executive or VP differs greatly from managing someone straight out of college.
GM CEO Mary Barra is on your board. She has been very deliberate in preparing the automaker to take on the AV market. How involved is she in Cruise's operations?
We work really, really closely with General Motors. We're developing vehicles together like the Origin. We're working on how we analyze safety in these vehicles and how we manufacture and deploy them at scale. But we are an independent company. We have a board of directors, she chairs it, and she's deeply invested in Cruise's success.
Get to know Vogt:
- In early 2020, he ran a marathon on seven continents in less than 82 hours. To train for it, he ran three marathons in 24 hours in three cities. "I have no desire to do anything like that again. It's tough," he says.
- Later that year, he and his wife opened Baia, a plant-based Italian restaurant in San Francisco.
- Vogt built his first self-driving car prototype at 13. It was a Power Wheels car with a webcam and a Pentium 233 computer in the middle, and it could follow yellow lines in the parking lot. He later built a more sophisticated AV prototype at MIT in the mid-2000s.
