The end result

If a mute person saw two people walking unknowingly towards a chasm which could mean certain death if either or both were to fall in, what would he do? It’s certainly a tough call — especially if they were walking at the same speed and were equidistant from the abyss.

Alan Winfield, professor of electronic engineering at the University of the West of England, Bristol, decided to also put the same question to what he calls an "ethical robot" to see what it would do. Accordingly, he used a robot which could predict the consequences of both its own actions, and the actions of other actors in its environment and programmed it to save other robots from coning to harm. He then designated an area as a virtual chasm and had two robots (as human proxies) walk towards it.

Approximately half the number of times in the trial runs the results were predictable: the ethical robot tried to save one of the "humans". However in the remaining number of times the results were surprising: the robot simply couldn’t make up its mind and kept dithering so much that in the end when there was no time left to save either, it simply wheeled around and walked away.

And there lies the biggest difference between machine ethics and the human variety for very few humans would walk away without saving at least one life every time they got the chance.

Sure, they would hesitate in making up their minds but finally with time running out they would reckon that one life saved is better than none. And the reason is we put an intrinsic value on human life which a machine, however sophisticated, doesn’t.