Given recent activity from companies like Google, Apple and Microsoft, it’s clear that the tech sector is driving straight into the heart of the auto industry.
Google, for example, recently announced the impending integration of its Android platform into some automobiles via a group called the Open Automotive Alliance (OAA). This partnership, which includes Google, Audi, GM, Honda, Hyundai and visual computing company NVIDIA, plans to roll out the first Android-integrated cars later this year, according to an OAA press release. The group’s self-stated mission is to establish “a common platform that will drive innovation and make technology in the car safer and more intuitive for everyone.”
Other software companies have gotten in on the action as well. Cars with Apple’s Siri integration began to appear in 2012. As stated in an article in TechnoBuffalo, Ford’s Sync AppLink, powered by Microsoft technology, is currently in 1.5 million cars in North America, with European and Asian roll-outs planned for later this year. Last month, the article noted, Ford announced four new apps for Sync AppLink, enabling drivers to “order pizzas for delivery or pick up, check in on your home or business, adjust your playlist, [and] find and pay for [a] parking space all with voice controls from a Ford vehicle.”
New software-powered safety features have also been on display over the past few years. Drivers can be warned when they are drifting out of a lane; they can get help with blind spots, and they can be alerted to step on the brake before hitting a car in front of them. The ultimate expression of this trend may be Google’s much-discussed self-driving car prototype, which the company first announced publicly in 2010 after it had already begun testing it on California roads.
“[Automobile] technology will advance faster than our ability to figure out the best way to use it.” –John Paul MacDuffie
In the OAA press release, Jen-Hsun Huang, president and chief executive officer of NVIDIA, stated: “The car is the ultimate mobile computer.” Which raises a number of questions: Do we want our cars to be the ultimate mobile computer? Are we willing to give up control behind the wheel? Are we ready to pay a higher sticker price? And are there overwhelming legal and ethical barriers to the world of driverless — or close-to-driverless — driving?
Automotive’s Brave New World — Or Is It?
Before considering what some think is a totally novel world of auto technology, says Wharton management professor John Paul MacDuffie, it’s worth noting that, in some ways, this is the latest chapter of a story that has been playing out for years. MacDuffie, who serves as director of the Program on Vehicle and Mobility Innovation (PVMI) at Wharton’s Mack Institute for Innovation Management, says there have been electronics in cars for a long time, but the amount of electronics — and the amount of software and hardware related to electronics and information technology — has lately been going steadily up.
Many consumers do not realize how much technology is already built into cars, says MacDuffie, pointing to basic functions, including steering, braking, suspension, fuel efficiency and emissions control, that all have a software component. In addition, some of the new technology features on the market seem less exotic when one realizes that, over the years, many of them have started as options and become standard. Anti-lock brakes and airbags are two examples.
As MacDuffie points out, it is not that there is software in cars; it is that there is a lot more software, and it is being developed more quickly. For some car buyers, it isn’t fast enough. “You sometimes hear people criticize the automakers for not jumping on putting phones in cars earlier, or doing it at a point when they could have persuaded most people to have their car phone be their source of mobile communication,” MacDuffie notes.
And apparently, when consumers first heard about Google’s self-driving car prototype, some wanted to get one right away, according to an article in Fast Company, which quoted Andrew Chatham, senior staff engineer and off-board software leader for Google’s self-driving car program: “One moment, the idea of a self-driving automobile was pure fantasy, something out of a science fiction film. The next, it was a reality, and people immediately wanted to know more about it and when they could buy one.”
Wharton marketing professor David Reibstein cites a variety of benefits from all the new safety and entertainment features being developed for cars. “I think [auto companies] are offering safer transportation, making it easier to drive. And it will be easier to get things done that people want to get done. There is a whole set of consumer benefits that will result.”
Cars and Computers — Irreconcilable Differences?
The benchmark for consumers is most likely the giddy pace of personal computing innovations — continually-updated smartphones, tablets and notebooks whose developers seem to anticipate consumers’ every need and desire. Such devices have quickly become an intuitive part of daily routines. Why, some might ask, can’t automakers be as agile as software companies? Why aren’t cars that drive themselves already available?
It’s not that easy, according to some of the experts. “There’s the fact that a vehicle [is] a big, heavy, fast-moving object that operates in the public space and therefore has consequences for public safety, which isn’t true for electronic gear,” says MacDuffie. With personal computing innovations, “you can have important parts of your productivity or your private data compromised, but it’s not like a death or an injury. And it mostly affects you, as opposed to other people.”
Technology, he adds, is a fast-moving industry that tends to operate according to open standards, with short product cycles and rapid obsolescence. The auto industry is the opposite — proprietary standards, long product cycles and products that stay in use a long time. Once a vehicle’s first owner is done with it, it is resold as a used car.
As a result, automakers may sometimes feel they have little more than a crystal ball to figure out what gadgets to incorporate. As James E. Lentz, chief executive of Toyota North America, noted in The New York Times, “I have got to guess today what you will want in a car 10 years from now. Today, I have got to decide what is going in a car in 2018, and that model will likely be sold until 2024.”
Plus, the risk inherent in incorporating the “wrong” features in a car can heavily impact the bottom line. MacDuffie paraphrases what Carlos Ghosn, the chief executive officer of Renault and Nissan, said at a recent conference: “If we put in a new technology that we understand consumers want and they find it hard to use, or they have multiple vehicles and don’t want to learn more than one system, they may go from potentially being pleased [with our vehicle] to thinking our vehicle is low-quality.” A lot of the interfaces in cars, MacDuffie says, “have been criticized pretty strongly across the board — from Mercedes to BMW to Ford — in the consumer reports and JD Power surveys.”
After all, it is hard to simply swap out built-in features in next year’s model when consumers are dissatisfied. “The idea that this could be very modular has been tried, and it runs into the difficulty of how integrated a car’s design is,” Macduffie notes, adding that anything incorporated into a car’s dashboard has to resist vibration and temperature extremes as well as fit in with, and work alongside, all the other equipment.
“I’ve got to guess today what you will want in a car 10 years from now.”–James E. Lentz
MacDuffie suggests that carmakers may achieve the most success going forward if they work to create more open standards — enabling consumers to bring in devices of their choice and use them — rather than trying to build in ever-more-flashy features. He returns to the example of car phones, an option that never really caught on. “Of course there were a few, but imagine if they had succeeded in installing them [across the board]. How long would it have been before they would have seemed impossibly big and clunky — no matter how well they worked at first — because they would have been there for the whole life of the vehicle?” He makes a similar point about installed navigation in cars, calling it a fairly high-end, expensive option “that always stays a step or two behind how much sophisticated GPS you can get either from a freestanding device, such as a Garmin or a TomTom, or now from your phone.”
What if software in cars could eventually be updated or repaired remotely so that consumers didn’t have to take their cars to the dealer? This might eliminate the obsolescence issue. Yet turning your car into a “network node” for software spawns its own set of privacy and security concerns, according to MacDuffie. “You certainly don’t want a hacker to be able to play games with your braking software.” He acknowledges, however, that automakers will want to integrate exciting features that will attract more tech-savvy or younger consumers and provide higher profits, as options or standard equipment, for the companies.
What Do Consumers Really Want?
The auto industry, Reibstein says, is only going to incorporate features that customers want and respond to. While he acknowledges the risk inherent in the industry’s long lead times, he points to the fact that the companies all do consumer testing before introducing new gadgets. “There are enough aspects of [the new features] that the companies could start trending toward and [learning about] even though you don’t have the complete product yet.”
Reibstein, who has been conducting an ongoing annual survey about self-driving cars for more than a decade, describes the trends he has seen. “Ten years ago, when I was doing [the survey], people were very skeptical. [One response was:] ‘I don’t want to give up having control of my car.’ I think some of that is dissipating. Now there are cars that will slow you down if you have it on cruise control and are getting too close to the car in front, and cars that help you park…. As we see that, people are getting more accustomed to some of the advantages and features the industry is working on.”
Despite all the appeal of hands-free, safer, productivity-enhancing and just plain “cool” automobile features, some consumers may run into a wall: the price. Although automakers these days are courting younger buyers in particular, many of these consumers cannot currently afford high-end cars or even new cars. According to The New York Times, which drew data from an IHS Automotive analysis, the percentage of new cars registered to buyers ages 18 to 34 has remained flat since 2009, hovering between 10% and 13%. The article also states that according to the global consulting firm Deloitte, which surveyed buyers born between 1977 and 1994, four out of five respondents said cost was the main barrier to owning or leasing a vehicle.
“There’s the fact of a vehicle being this big, heavy, fast-moving object that operates in the public space and therefore has consequences for public safety, which isn’t true for [personal] electronic gear.” –John Paul MacDuffie
In addition, some consumers may be daunted by all the bells and whistles, in spite of the fact that others were clamoring for a self-driving car the minute the announcement came out. According to Bryan Reimer, a research scientist in the Massachusetts Institute of Technology AgeLab and associate director of The New England University Transportation Center, there is a delicate balance between providing drivers with a safe and effective source of information versus overwhelming them with too much information. “While some systems may try to address too many features, there are numerous examples of clear intuitive designs that provide drivers with a good balance. Consumers clearly favor vehicles with easy to use … systems. This should motivate manufacturers to meet the consumer demand in a safe and effective interface.”
Legal and Ethical Questions
Are we really headed toward mass-produced self-driving cars, and if so, when?
“We will see some penetration in five to six years,” Reibstein predicts. But for mass production, he leans toward “easily 10 years.” As for demographics, he suggests that the majority of customers will be people with a long commute who want to use the time more productively. Reibstein, himself, says he might like a self-driving car “because right now, when I’m driving I’m not texting or doing email or other productive things.” If safety is perceived as a strong benefit of the product, he adds, it will appeal to families. Affluent buyers are also likely to be early adopters because of the cost.
“The technology will advance faster than our ability to figure out the best way to use it,” MacDuffie states, adding that the main barriers to bringing autonomous driving to the market will most likely be legal and ethical concerns. “Once you have a set of software algorithms making choices about whether or not to swerve in a given situation, it’s probably [focused on] maximizing the survival of the driver. But what does that mean about animals or property, for example. How close are we to being ready to turn those decisions over to technology?”
And if your self-driving car has an accident, who is liable — you or the company that wrote the software? “We do have the precedent of auto-pilot features being used in airplanes,” MacDuffie notes, but these are tightly regulated by the FAA.
According to Reimer, there is a mix of drivers who really do want self-driving cars and others who enjoy driving. “For the foreseeable future, I think we will see an increase in the level of automated vehicle features, but portions of the driving population will still prefer the option of ‘manual’ control.” There are clear safety advantages to automation, he adds, but there are also many “potential unintended consequences” that he says will call for new research and innovation in the driver-vehicle interface.
Reimer’s research at MIT is focused on understanding how drivers respond to the increasing complexity of the operating environment. In an article in Fast Company, he commented: “Approximately 80% of people we polled boasted about their prowess behind the wheel.” He suggests that people’s pride in their own driving ability — justified or not — will make it hard for them to accept “how technology is going to change the relationship between machine and operator.”
Reibstein predicts that when the self-driving car is finally launched, there will inevitably be problems and bugs that will get “lots of notoriety and publicity,” but he says the likelihood is that overall driving errors will be reduced relative to human error. He notes the experience of the Google self-driving car road tests. The fleet of about 10 autonomous cars has reportedly experienced two accidents so far. One was when a car was being driven manually; the other occurred when someone who was not part of the driving test accidentally drove into one of the self-driving cars.