The three laws of robotics, according to science fiction author Isaac Asimov, are:
1. A robot may not injure a human being or, through inaction, allow a human being to come to harm.
2. A robot must obey the orders given to it by human beings, except where such orders would conflict with the First Law.
3. A robot must protect its own existence as long as such protection does not conflict with the First or Second Law.
I would gladly have accepted a $20 million Pentagon contract for the job of pointing out these three laws.
OK, maybe $25 million.
Sadly, the Pentagon has instead hired a bunch of philosophy professors from leading U.S. universities to tell them how to make robots murder people morally and ethically.
Of course, this conflicts with the first law above. A robot designed to kill human beings is designed to violate the first law.
The whole project even more fundamentally violates the second law. The Pentagon is designing robots to obey orders precisely when they violate the first law, and to always obey orders without any exception. That’s the advantage of using a robot. The advantage is not in risking the well-being of a robot instead of a soldier. The Pentagon doesn’t care about that, except in certain situations in which too many deaths of its own humans create political difficulties. And there are just as many situations in which there are political advantages for the Pentagon in losing its own human lives: “The sacrifice of American lives is a crucial step in the ritual of commitment,” wrote William P. Bundy of the CIA, an advisor to Presidents Kennedy and Johnson. A moral being would disobey the orders these robots are being designed to carry-out, and — by being robots — to carry out without any question of refusal. Only a U.S. philosophy professor could imagine applying a varnish of “morality” to this project.
The Third Law should be a warning to us. Having tossed aside Laws one and two, what limitations are left to be applied should Law three be implemented? Assume the Pentagon designs its robots to protect their own existence, except when . . . what? Except when doing so would require disobeying a more important order? But which order is more important? Except when doing so would require killing the wrong kind of person(s)? But which are they? The humans not threatening the robot? That’s rather a failure as a limitation.
Let’s face it, the Pentagon needs brand new laws of robotics. May I suggest the following:
1. A Pentagon robot must kill and injure human beings as ordered.
2. A Pentagon robot must obey all orders, except where such orders result from human weakness and conflict with the mission to kill and injure.
3. A Pentagon robot must protect its own existence as long as such protection does not conflict with the First or Second Law.
This set of laws differs from Asimov’s in a number of ways. For one thing, it completely lacks morality. It is designed for killing, not protecting. By prioritizing killing in the First Law, rather than protecting, this set of laws also allows for the possibility of robots sacrificing themselves to kill rather than to protect — as well as the possibility of robots turning on their masters.
This set of laws differs much less — possibly not at all — from the set of laws currently followed by human members of the U.S. military. The great distinction that people imagine between autonomous and piloted drones vanishes when you learn a little about the thought habits of human drone pilots. They, like other members of the U.S. military, follow these laws:
1. A Pentagon human must kill and injure human beings as ordered.
2. A Pentagon human must obey all orders, except where such orders result from human weakness and conflict with the mission to kill and injure.
3. A Pentagon human must protect its own existence as long as such protection does not conflict with the First or Second Law.
The job of the philosophy professors is to apply these laws to robots while neither changing them nor letting on to have figured out what they are. In other words, it’s just like teaching a course in the classics to a room full of students. Thank goodness our academia has produced the men and women for this job.