How AI is taking controls of our fate

Would you trade the chore of staying focused on the road in exchange for a thirty-minute nap? What if I told you that the condition for that tempting change is putting your life in trust of an advanced computer?
Who will be to blame if you happen to get injured in an autonomous car? I will try to answer those questions today and invite you to join me in this discussion.

It is no secret that driving a car can be tiring, be it in traffic or on empty stretches of a highway, and energy drinks can dismiss the need for your inevitable nap oh so slightly before you just have to stop and rest.
Elon Musk has an answer for this problem, and you can get a glimpse of it today, with some obvious limitations that many are too comfortable looking past already.

Photo — Tesla

What I’m talking about is Tesla’s Autopilot, which, for now, is closer to BMW’s Driving Assistant than to a fully autonomous system, but a quick web search reveals that, despite the manufacturer’s recommendations, Tesla owners are already getting from point A to point B without much driver input. Musk hopes that truly self-driving cars will be available in 5 to 10 years, and no matter how successful his vehicles are at predicting and avoiding accidents, and staying in lane, today, full responsibility is still that of the person sitting behind the steering wheel. There are, however, plenty of videos online that show Tesla owners sleeping in the driver seat while on a highway, which would indicate the responsibility, at least in their own understanding, is no longer on them.

Ok, great, this is all fine and dandy, but what happens when an accident takes place? Computers can’t improvise, there will always be some failure rate. Is it worth trusting with your life in exchange for a convenient drive to the airport?What about the life of other people on the road or next to it?

The trolley problem is a thought experiment that aims to determine human behavior in the light of an extreme moral dilemma. To save five lives and sacrifice one, or to let the events run their course and not be involved. With cars having to calculate the safest way out of an accident, such scenario needs to be accounted for. The question is not whether we are ready to risk our own lives to drive autonomous cars, but rather if we’re willing to risk the lives of others. After all, being in traffic is not only about ourselves, but also about everyone else around. When the advanced technology enables the drivers to turn into passive commuters the choice of which lane to drive, what obstacles to avoid, and ultimately, which life to save, is carried over to the engineers behind algorithms controlling the car. The moral dilemma is a problematic issue and the main cause for varying regulations concerning autonomous driving all around the world.

India approached self-driving vehicles from a different perspective, backing its decision to ban those with the intention to prevent loss of jobs that would otherwise be taken away from truck or bus drivers. No matter if you look at that matter form economic or ethical point of view, the world doesn’t appear to be quite ready for replacing drivers with computers.

Nobody can deny that Tesla’s Autopilot is an excellent feat of engineering, but the heated debate is justified. Hundreds of success stories about this system effectively preventing collisions cannot diminish the consequences of any hardware failure of software errors that might potentially occur.

What happens if the system misinterprets the danger, or lack thereof, and results in a fatality? The Autopilot will not stop the driver from collision if they apply the accelerator pedal, and while Tesla say such behavior is meant not to undermine the driver’s choices in an emergency, at times overriding human’s control could lead to a more favorable outcome. The moral dilemmas are countless when designing a self-driving system. Should we eliminate the human error from the equation completely? Would fully autonomous road traffic be safer than the one constituting of people only?

Izabela Błażowska co-founder of ChallengeRocket.com, a global platform specializing in the use of AI in automated recruitment in IT says:

Artificial Intelligence will certainly improve our lives in countless areas, but the reluctance towards automated solutions being a replacement for humans performing some activities is very apparent today. Change never happens overnight, but I strongly believe the sooner we as a society start introducing such solutions, the more and quicker we can benefit. We can be certain that this new era in automotive industry is emerging right before our eyes.”

What about you? Every opinion on this subject is equally as valuable. Is it reasonable to fully commit to AI-controlled cars? For now, this debate remains open.

The views and opinions expressed in this article are those of the authors and do not necessarily reflect the official policy or position of Tesla.