This year’s Geektime Techfest hosted a special panel yesterday, discussing what to watch as autonomous vehicles continue their invasion of the market. The trio was asked a slew of questions, providing more insight to the buzz about self-driving cars’ perceived inevitable overthrow of today’s consumer car industry. While the panel seemed to take that inevitability as a given, they begrudgingly admitted some things that would hold back that eventuality.
“I think enabling all of us to be in self-driving cars will take longer because of regulation,” said Emanuel Timor, general partner at Vertex Venture Capital. “In 10-15 years, what will be the impact of the Ubers of the world? The impact of moving to current ownership to Uber model? I think the technology will be there earlier, but mass acceptance by regular drivers will take longer.”
There are a few reasons for the delay besides regulation. In fact, the US government has thrown support behind the notion already of shifting to autonomous autos, saying they likely would make roads safer.
However taking a step back, it looks more likely that the issue lies in consumer habits.
Changing the way we see cars
There is freedom having your own wheels and human foot on the pedal that is still very much embedded in our mindset. As such the industry will need to work hard to encourage this changeover to a new system.
“We should change our language. We won’t be driving autonomous cars, we’ll be using autonomous cars,” emphasized Dr. Ran Gazit, GM Israel’s team leader for smart sensing and vision systems. “Unlike today where each one of us will have our own cars, they will be owned by someone else.”
“We see a tendency in the world … people in New York City don’t want to own a car. Young people don’t want private cars anymore because they can get anywhere using public transportation, or if they need to then they can just rent it,” Gazit argued.
Gazit also suggested that decongestion and undoing pollution were incentives to move from private ownership to collective ownership and the development of new mobility services, which he credits his own company with pursuing through its relationship with Lyft.
“People are willing to let go of the comfort,” asserted Innoviz Co-Founder and CEO Omer Keilaf. “They just want to go from Point A to Point B in the most efficient way, with the lowest cost.”
Cars transitioning to a service economy
But there’s another issue that comes up in Gazit’s words: farming out vehicles as a service. Much has been discussed in that regard, as is the case with GM and Lyft, but making a transition to that sort of economy means something else for traditional product-only companies.
Timor framed the question, “Can the manufacturers also become a service provider? That’s a major challenge. Most companies can’t evolve from being a hardware company to dealing directly with the customers.”
Manufacturers are also focused on how to construct cars in what many predict will be a significant but relatively short period of transition between human-driven cars and autonomous vehicles.
That will lead to a long period of coexistence between human drivers and autonomous vehicles on the road.
“I think it’ll go a different way. I think they will shift production lines from direction where some cars won’t be fully autonomous,” suggested Keilaf, “then after some time when those cars prove themselves, then they will stop producing human-capable driving.”
That will likely shake up the used car industry, as society grapples with the shift.
“We can start thinking about autonomous-ready cars that will have a drive-by-wire capability,” Gazit chimed in. “When you press a pedal brake, a computer does the brake for you. Very few do that today, but those are the cars that can be retrofitted.”
We are already seeing discussion about retrofitting, but trying to upgrade a 1995 Camaro to fit in with the autonomous age probably makes no sense. Even cars that lend themselves better to retrofitting will face challenges.
“It would be much cheaper to buy the car,” said Timor. “It means a lot of sensors, a lot of computing power” to fundamentally flip the car over into something robotic. Only a few cars on the market will lend themselves to a self-driving upgrade later, mainly cars that have some sort of computer integration already.
“If you say a specific organization or university wants to do retrofitting, that might make sense, but for a regular consumer it won’t make sense. You’d see, it wouldn’t be embedded in the car.”
Why a specific organization? Gazit suggested persuading corporations or universities to roll out on-campus use of self-driving vehicles, perhaps exclusively, in order to encourage further adoption.
“In terms of the technology today that will give insight answering the problem – the tech we have today can’t drive from anywhere to anywhere,” Gazit said, saying that areas of Pittsburgh had to be mapped judiciously for Uber’s technology to work. That process, dubbed geofencing, means that “Uber’s autonomous cars couldn’t be brought over to Tel Aviv and expected to work. Not even one meter. The mapping has to be very careful.”
Until someone or some group decides to give more of a sandbox for autonomous cars to play in, we will still have barriers to mass adoption.
“Of course we would like to see autonomous cars that can drive anywhere under all conditions, but it will take time to get to that stage.”
Timor concluded with a more optimistic parting shot, though.
“I think there’s a huge opportunity now. If you look at the last decade there was already [a] $250 million value creation if you look at the leaders in shared mobility, which came from revolutionizing the taxi industry. Turns out three years later they may change the entire automotive industry.”