I, Robot; The Robot Series

Isaac Asimov

Bantam Doubleday Dell Publishing Group, Inc.

New York, NY

1950 & 1991

Reviewed by: Jerry Tapiador

Asimov accomplishes this in a very elegant and semi-linear fashion. He previews the book in the introduction with a character by the name of Susan Calvin. Her name isn’t important but what is important is her occupation. She is a "Robopsychologist (viii)," which right away brings up this notion of strong AI. However, the book doesn’t jump straight into a machine so complex and with such reasoning abilities that it would require a psychologist.

Instead it starts in flashback fashion at the first humanlike robot. The robot can’t talk and it can’t reason (weak AI). It is programmed to take care of a child. The robot itself is not interesting, except for comical reasons in which it is an extrapolation of what was known in the 50’s. The interesting part was the cultural developments that occurred. The paranoia of robots that swept the countryside the prejudices that occurred and the clinging to old fashion solutions.

After the first chapter is concluded the book follows the development of robots from this programmed non-reasoning thing to a reasoning entity (strong AI) with three rules imbedded into it. The rules are the famed "Laws of Robotics."

They are mentioned throughout the book and cause problems with the new developments of robots. The rules are kept in the robots for safety purposes and the logic problems that arise from these rules and tasks were frequently subverted by command rather than reprogramming or restructuring the laws.

The majority of the book addresses the strong AI concepts and the problems that might occur. The book is especially good at addressing items such as faulty logic, logical fallacies, and circular logic. These are all problems that occur when the robots start to think like humans. In one particular chapter the faulty logic goes so extreme that one of the robots seizes power and starts a cult on a power relay station (Ch 3 Reason).

The end of the book is where the story line climaxes as well as the AI. In the end, a fourth law is drawn from the first three laws. However, It is ranked with a higher priority since logically it is of a higher order. It is the Zeroth Law of Robotics that states: "A robot may not injure humanity, or, through inaction, allow humanity to come to harm (pg269)." It is at that point that the robots effectively take over and watch over humans not in a sinister apocalyptic way but in a motherly caring way.

The AI aspects of the book permeate throughout. It is both in the human and in the robotic characters. The book starts with weak AI, an easily programmed but limited robot, and proceeds to strong AI, a robot with reason but difficult to control. It culminates in a sort of hyper-AI in that the ‘Machine’ is in a way omniscient. I consider it omniscient since it understand relationships and things that we do not. Furthermore, unlike us it somehow has the ability to reject wrong data. I admit that the Enquirer is not the most credible source but that isn’t the same as rejecting ALL wrong data. I’m skeptical of whether this is possible, but then again this is science-fiction.

Other AI topics were culturally based. Ideas such as philosophical and religious implications were brought up as well. Also, the use of power positions was very interesting. In the start of the book, humans were very blatantly the ‘boss.’ Then later the ‘Machine,’ in a very cunning manner, basically became the boss.