The true flaw of driverless cars isn’t the tech

My three-year-old daughter, Carson, is perhaps Waymo's biggest fan. Waymo vehicles, which grew out of Google's self-driving project, are tested so frequently in our San Francisco neighborhood that we typically see four of them on our 15-minute dog walk . Carson and I play Find the Waymo and sing about the laser sensors that allow the vehicle to see. "The lidar on the Waymo is doing pulse, pulse, pulse..."

Carson likes Waymo because they look distinctive, their 29 cameras, five lidar and six radar sensors protruding on all sides of a Jaguar I-Pace model. However, she is not impressed that a robot is driving. That's just the reality she was born into. It's no more special than on-demand Bluey episodes.

Former Waymo CEO John Krafcik told me last year that Carson will never need a driver's license. "She'll be able to use a Waymo almost anywhere she is," he said. "So that's one less thing to worry about."

In 2014, Travis Kalanick said the prospect of driverless cars posed an existential challenge for Uber... Since then, it's become clear that the economics are more complex

Since then it has become clear that economics is more complex

Thirteen years is a long way off, but the wheels are already in motion for San Francisco to become the first major city with a driverless ride-hailing service. Last summer, Waymo and GM-owned Cruise received approval to "receive compensation for autonomous services offered to the public." Neither group has significant revenues today, but investors value each over $30 billion -- double that of Lyft, which operates in more than 600 cities in all 50 states.

Investors are betting on the belief that once these vehicles prove they are safer than humans, they will turn into ATMs on wheels. This vision has attracted the world's largest companies, including Apple, Microsoft and Amazon.

Nevertheless, I remain skeptical. It's not so much the technique. When I used Waymo to pick up my family for a 20-minute test drive, the robot made no noticeable mistakes. It's assertive and smooth. And the "true driverless" ride-hailing service Waymo, which Waymo has been operating in Phoenix for more than two years, has not been involved in any major incidents. No, the reason these services are not widely available is mainly due to the lack of a clear business model.

Back in 2014, Travis Kalanick, then CEO of Uber, said the prospect of driverless cars posed an existential challenge for Uber. Human drivers make up two-thirds of the cost of an Uber ride, he argued, "so if there's no other guy in the car, the cost of an Uber ride is cheaper."

Since then it has become clear that economics is more complex. What made Uber so disruptive is the gig economy. The app serves nearly 100 million people in more than 10,000 cities every month without Uber owning a single car. Its drivers are contractors, so they only get paid when they carry passengers. For example, if the driver is waiting at the airport, that's not Uber's problem. It has sophisticated software to match riders to riders and to do it quickly, but endures few penalties for having too many riders at once or too few.

The driverless industry throws away these benefits. Instead of hiring drivers cheaply, Waymo has to test for years in any given market. Instead of the lucrative, asset-lite business model, Cruise instead has to build or buy vehicles that cost about $200,000 each. Put in too many cars and they waste capital. Too few and they will frustrate passengers with long waits.

A driver also plays the implied role of conductor, making sure the car has gas and discouraging illegal activity. Driverless cars need to be maintained and monitored to do the same. "Driverless doesn't mean peopleless," says Ashley Nunes, a researcher at MIT. His work on the subject concluded that robo-taxis cannot compete in price with the gig economy. “Robotaxis replace one set of human costs, the human driver, with another, inefficiency,” he adds.

Like the industry, I am confident that the technical challenges faced by robotaxis can be solved. But the business model seems flawed. Far more plausible are the autonomous tractors that Deere built to roam empty fields during seasonal work. Autonomous trucks also seem to have a great future.

When robotic taxis finally open to the public in San Francisco, they risk becoming a gimmick. It's fun to try, sure, but they cost the same as an Uber, don't get there faster, and refuse to speed if you're in a hurry. A lot could change by 2035, when my daughter reaches driving age. But even so, I may still have to join previous generations of parents in enjoying the unique horror of sliding into the passenger seat.

Patrick McGee is FT's San Francisco correspondent

continue reading

https://dailytechnonewsllc.com/the-true-flaw-of-driverless-cars-isnt-the-tech/