Technological progress is not a phenomenon only of our times. It is as old as man. However much of a break with the past today’s technologies—robots, computers, artificial intelligence—may seem, they are in fact part of a continuum
Pro: We All Get More Time
One study on passenger car travel in the United States reported that Americans spend collectively 175 million hours a year in traffic. All of that time could be used more productively with an autonomous vehicle. 175 million hours of business time regained will be reflected in the national economy. In the book Sustainability prospects for autonomous vehicles : Environmental, social, and urban it reveals how bad traffic is really getting “Another measure of the growth in congestion is a slowdown in the speed of urban traffic, as indicated by a study of London, Los Angeles, Paris, and Stuttgart (INRIX 2014). In London it was 21 miles per hours (mph) in 2013 and was projected to fall to 16 mph in 2030; in Los Angeles, from 21 to 15 mph.”
Con: Less Thinking Time
Safe driving practices, allow you to focus on one thing and get away from screens, from work, and much more. For some, driving is the only time in the day to think clearly, to meditate, or to just sit still. This can still be done in an autonomous car obviously, but you’ll no longer be forced to focus on driving.
Pro: No More Typical Car Costs
This one is not entirely for certain, however many speculators believe that in the world of autonomous cars companies and manufacturers will begin to take on the liability for their autonomous cars. Similar to how things work today with Uber, Lyft, and the like. When you don’t own a personal car anymore you don’t need to insure it.
The same goes for other things you typically pay for when owning a car. Goodbye oil changes, checkups, tire rotations, new tires, dead batteries, new transmissions, all of it. With autonomous cars, companies will have to absorb these costs.
Con: New Things to Pay For
Because autonomous cars are more efficient, and people will be carpooling, the actual cost of transportation per person will go down. You won't be directly paying for auto insurance, gasoline, annual maintenance, etc. but all of those costs will be reflected in the price of the subscription to use an autonomous car.
A few other costs that will arise will be the complete reconstruction of our nation’s current infrastructure. For us to really benefit from AVs (autonomous vehicles) we’ll need smart streets, smart street lamps, highways, freeways, buildings, etc. Who will pay for this?
Pro: Traffic Safety
Possibly one of the biggest benefits of autonomous vehicles is traffic safety. About 90% of motor vehicle accidents are caused by a human error. About a third of those involved a distracted driver. In a fully “smart” city the road would be talking to the car and vice versa making it very statistically improbable for accidents to happen. It also just generally means less traffic.
Attached to this benefit is the cost on city, and state governments that traffic incidents are. During the COVID-19 pandemic. From the start of shelter-in-place (about a month) California saved over $40 million a day or roughly $1 billion altogether. This can be looked at as a test-run for what the autonomous future will be. Obviously there’s a difference between the whole state shutdown and self driving cars being fully adopted but the numbers show that without people behind the wheel accidents are down.
Con: Some Things Are Gonna Get Hit
Many speculators talk about how long it will take for us to actually become fully autonomous. Mckinsey Consulting Firm marks this process in phases:
Era 1 (2015 - 2025): Fully autonomous vehicles developed for consumers
Era 2 (2025 - 2040): Consumers begin to adopt AVs
Era 3 (2040-2050~): AVs become the primary means of transport
Their evaluation of the phases breaks down what each of those entail. The era of concern is Era 2 which we’re coming to pretty quickly. Era 2 is the time when some AVs appear on roads and have to drive compatibly with human drivers. We’ve already got a small taste of what this will be like with practically self-driving Teslas. In order for these cars to become safe, they’ve got to be unsafe for a while. Like a baby learning to walk, it’s gotta fall down a few times. Unlike the simple coding of an ordinary robot, autonomous technology requires machine learning. Emphasis on the learning, which comes from experience. For a car to recognize a kid flying by on a skateboard, the engineers will show the car thousands of pictures of kids on skateboards, kids approaching cars on skateboards, etc. But the only way the car will really learn is if it has a few times messing up and actually hitting the kid; which obviously is not ok. There’s a Bloomberg article by Zachary R. Mider called “You Gotta Smash a Few Cars to Make an Autonomous Vehicle” , Zachary details why this is the case.
“The more experiences they have, the smarter these machines get. That’s part of the problem, Kalra argues, with keeping autonomous cars in a lab until they’re perfect. If we
really wanted to maximize total lives saved, she says, we might even put autonomous cars on the road while they’re still more dangerous than humans, to speed up their education.
Even if we build a perfect driverless car, how will we know it? The only way to be certain would be to put it on the road. But since fatal accidents are statistically rare—in the U.S., about one for every 86 million miles traveled—the amount of necessary testing would be mind-boggling. In another Rand paper, Kalra estimates an autonomous car would have to travel 275 million failure-free miles to prove itself no more deadly than a human driver, a distance that would take 100 test cars more than 12 years of nonstop driving to cover.”
Since 2016 there have only been six autonomous vehicle fatalities on record. Which is a pretty good track record, considering as of February of this year Tesla announced their cars have driven 3 billion miles on autopilot. All of those miles have been collected as data points, and learnings for Tesla’s autonomous systems. Another reason why Tesla is literal miles ahead of the competition. Perhaps one of the smartest things Mr. Musk has done is put these cars on the road with Autopilot as opposed to spending years of testing and refining.