What controls are needed on robots?

I watch with fascination the TV commercials that show a Ford automobile parallel parking on a street.
I remember how many practice hours I spent learning to put my father’s ’65 Chrysler into a tight space.
We have seen robot vacuum cleaners. We have played with robot toys and marveled at C3PO in “Star Wars,” who was human-like.
Today we are closing in on having robots become part of our every day lives.
Perhaps the first robots were the metal adding machines. As long as you punched in the numbers correctly, the column being added was correct. It was far faster than mentally adding each column.
Later, electronic calculators sped up the process. And as Texas Instruments added more functions, they made the work of engineers and students much easier.
They truly were a first generation of a robot.
When we began putting humans into space, much of the work was programmed into very small computers by today’s standards. The first astronauts were more hands-on.
Today, everything is controlled by robotic functions.
When you strap yourself into a seat in one of our modern airplanes, the pilot’s function is to taxi the plane to the take-off runway and then, when the flight has landed, to taxi it to the gate.
The take-off, flight, and landing all is handled by robotic computers that constantly monitor everything that’s happening. Similarly, when you start your car engine, an automatic robotic system check does an inspection of the vehicle.
Robots are making their way into our every-day lives.
We send robot submarines to the deepest parts of the ocean to explore and discover new animal species. We might control them from the surface, but the real machine is far below.
Through war, we have created a whole new set of robots to spy, carry, infiltrate, and make war on enemies. From a distant country, we can order a drone flying some 11 miles high to attack a home or vehicle.
We can operate trains without an engineer being aboard and send it across the country. Google has perfected a car that can travel up and down streets photographing everything it sees in 360 degrees.
But in the future, as robots take on more human responsibilities, how are we going to program them to operate with a set of morals and ethics.
Today, if a truck is travelling down a highway and the driver sees a group of five people standing and blocking the truck’s path, and a single person in the oncoming lane, and is unable to stop in time, the driver is faced with a difficult decision.
In the future, how would you develop the logic, ethics, and morals for a robot to handle that decision?
In war, a drone may hover and realize that an enemy is in a home, and may suspect that innocent men, women, and children also are in that home, the ethical dilemma for the drone will be whether or not to attack.
A robot soldier not controlled by a human would have to make a decision about whether the person they are facing is an enemy soldier or a terrified civilian.
We will have to think about how we make them more rational and clear-thinking than ourselves. What qualities do we want these future robots to have and use? What moral judgments do we want them to make?
Will we want to put restrictions on their use?
Successful technologies are quickly adopted. And robots will become more a part of our every-day life.
We just have to think about what controls we wish to place on them.

Facebooktwitterredditpinterestlinkedinmail