When it comes to technological progress, driverless vehicles are riding in the fast lane. Yet there are still many questions about the ethics of such technology that need to be answered before these self-driving cars will be able to hit the highways.
“If there’s an accident involving a driverless car, what policies do we have in place? What kind of insurance coverage do they have? And who needs to take insurance?” asks Carnegie Mellon President Subra Suresh. The University recently received a $10 million grant to study the ethical issues of artificial intelligence, including self-driving cars.
Those questions are also on the minds of insurers themselves, who want to know who might be considered “at fault” in the event of a driverless vehicle accident. Some experts suggest that accidents might become cases of product liability, eliminating the need for personal driver insurance altogether.
The manufacturers of self-driving vehicles — including Tesla, which just released a video of how its cars “see” the world — argue that roads will be much safer with driverless vehicles, eliminating the risks of drunk driving, low visibility, or distracted driving. Motor vehicle accidents were the leading cause of death in 2014 among young people ages 16 through 24, and DUI charges can leave drivers with a permanent criminal record that may negatively impact their chances of employment, loans, or other opportunities. Without the risk of human error, these circumstances might become mere problems of the past.
The current technology is not perfect, however. Tesla’s new video gives an impressive example of how the car, capturing and reacting to the world around it at a rate of 2100 frames per second, can safely navigate around two joggers on the side of the road. It doesn’t, however, answer the question of what the car might do when faced with a do-or-die decision of, for example, either hitting the joggers or crashing into a ditch and injuring the driver.
These questions partially explain why a full adoption of driverless technologies might be further off than some originally thought, despite their rapid progress.
“In 2019 you will be able to buy a car with an autopilot system where you can take your hands off the wheel for up to three minutes,” said insurance industry researcher Matthew Avery. “When the customer can buy a car like this it’s going to be confusing to him to understand what is allowed. Do I have to be in control or can I get in the back of the car and have a sleep?”
Avery predicts that consumers will have to wait until the mid-2020s before fully autonomous vehicles hit the roadways. But with Tesla’s October announcement that newer models of their cars will already have the software for driverless navigation built-in, the final switch-over may hinge more on government policy than technological capability.