The Biggest Obstacles to Our Driverless Future

Transcript

We've been closely tracking the progress of autonomous vehicles in Trends for our subscribers since at least 2007. Now, nearly a decade later, the technology is finally catching up to the vision-thanks to rapid advances in artificial intelligence, sensing systems, and processing power-and driverless cars are nearly ready to take over the road.

Unfortunately, technologies often move faster than the society that ultimately accepts or rejects them. In this case, the benefits of driverless cars couldn't be clearer.

• $488 billion from preventing crashes
• $507 billion from improving productivity
• $158 billion from saving fuel
• $138 billion in additional productivity improvements from avoiding traffic jams
• $11 billion in additional fuel savings from avoiding congestion

By taking the steering wheel away from human drivers, autonomous cars could save tens of thousands of lives each year. According to the National Highway Traffic Safety Administration, human mistakes account for 94 percent of car crashes, which cause about 38,000 fatalities per year.

Driverless vehicles would also enable people who can't drive, such as disabled or elderly people, the freedom to go anywhere in their own car.

• According to University of Texas research, every shared autonomous vehicle could take the place of up to nine privately owned vehicles. The average wait time for a car would be only a minute, and during the busiest times the wait would be a mere four minutes.
• A Columbia University study found that shared autonomous vehicles would reduce consumers' transportation costs by 75 percent.
• According to a Lawrence Berkeley National Laboratory study published in Nature Climate Change, greenhouse gas emissions of a battery-powered shared driverless car would be about 90 percent lower in 2030 than those of a 2014 gasoline-powered privately owned vehicle.
• Replacing private ownership of cars with shared autonomous cars would eliminate much of the current need for parking. Because cars are only in use about 4 percent of the time today, they must be parked for about twenty-three hours per day. But if they could stay in constant motion, 3,000 square miles of land that is currently used for parking lots and garages could be converted to more productive uses.

1. Legal liability
2. Consumer acceptance
3. Meddling by policymakers

The issue of legal liability is tricky when the car drives itself. Insurance companies haven't yet provided a solution. It's possible that the owner of the vehicle will be liable for any collisions. But this opens the possibility for lawsuits, as vehicle owners could reasonably claim that the vehicle manufacturer or the software firm that developed the navigation system had more responsibility for the car's actions than its owner.

Partly for this reason, but for others as well, consumer acceptance may be slow, initially. A Harris Interactive poll found that 59 percent of Americans are worried about who would be liable for car accidents.

Meanwhile, 52 percent are afraid that hackers will access the car's software and cause it to stop or to crash. And 33 percent dislike the potential for data about their movements being used by advertisers, insurance firms, or the government.

Another study, by the American Automobile Association, found that three of out of four Americans are afraid to entrust their safety to a driverless car. The fear is greater among women, at 81 percent, than men, at 67 percent. And more Baby Boomers are afraid of driverless cars, at 82 percent, than are their children or grandchildren, at 69 percent.

Policymakers share these same fears. At the state level, twenty-three states have drafted fifty-three bills to regulate autonomous cars. Five states have already enacted regulations on driverless vehicles-but all of them are different, which could lead to chaos as manufacturers will have to meet different standards in different places.

Despite objections from Google, California plans to require that every driverless car on its roads must have a steering wheel, brake and gas pedals, and a licensed driver in the front seat ready to take control of the car in an emergency.

Even in states that are trying to welcome the technology, existing regulations can have unexpected consequences. For example, Utah discovered that autonomous vehicles that follow each other closely and safely would violate an existing state law against tailgating. In 2015, the state had to change its law to exclude driverless cars.

The best solution would be for the federal government to create a small set of reasonable regulations to cover the entire country. U.S. Transportation Secretary Anthony Foxx has promised to create a set of rules for the operation of autonomous vehicles by this summer.

The National Highway Traffic Safety Administration has expressed a willingness to consider a driverless car's self-driving system as the "driver" of the vehicle, rather than the human who rides in it. In its response, the agency noted, "If no human occupant of the vehicle can actually drive the vehicle, it is more reasonable to identify the 'driver' as whatever (as opposed to whoever) is doing the driving."

Based on this trend, we offer the following forecasts:

First, completely autonomous cars will be on the road by the end of this decade.

Second, the best policy approach for the federal government would be to set as few regulations as possible until the technology evolves.

Other than laws that explicitly protect Americans' safety, policymakers should avoid taking actions that would serve only to delay the rollout of the autonomous vehicle revolution. History shows how damaging the slow regulatory process can be to the U.S. economy.

For example, a National Economic Research Associates study found that if the Federal Communications Commission had started issuing cellular licenses in 1972-instead of delaying the introduction of mobile phone technology for eleven years as it crafted regulations until 1983-the U.S. economy would have gained $86 billion, or $205 billion after adjusting for inflation.

Third, consumer acceptance for autonomous vehicles is likely to increase quickly as Americans grow accustomed to letting cars take over more and more of the driving that humans do today.

The American Automobile Association survey, mentioned earlier, which found that 75 percent of Americans are fearful of self-driving cars also revealed that a majority of them are actually eager to use semi-autonomous features in their next cars. These features include adaptive cruise control, lane-departure warning, automatic emergency braking, and self-parking technology.

The transition from routinely relying on these semi-autonomous features to entrusting a fully autonomous car will be a small one. As for fears about loss of privacy, those always disappear in a tradeoff for convenience. Consider that a survey found that 82 percent of Americans worried that Uber would not protect their personal data. Despite that uneasiness, the company doubled its business over the past year. Also, Americans are simply not driving as much as they once did.

A study at the University of Michigan Transportation Research Institute discovered that, in every age group, the percentage of Americans with a driver's license dropped from 2011 to 2014. Most critically, the percentage of 16-year-olds who got a driver's license plummeted by nearly half-from 46 percent in 1983 to 24 percent in 2014.

From shared bicycle services to ride-hailing apps like Uber and Lyft, younger Americans are increasing moving away from a mindset of ownership to one of sharing, so they are perfectly suited for a future in which cars will be shared and autonomous.

Comments