Artificial Intelligence: How Machines Think

by F. David Peat

Discusses the evolution of computers, problem solving logic, expert systems, natural language systems, robots, and computers that learn

356 pages, Paperback

First published January 1, 1985

About the author

He has worked actively as a theoretical physicist in England and Canada.

But Peat's interests expanded to include psychology, particularly that of Carl Jung, art and general aspects of culture, including that of Native America. Peat is the author of many books including a biography of David Bohm, with whom Peat collaborated, books on quantum theory and chaos theory, as well as a study of Synchronicity. Since moving to the village of Pari in Italy, Peat has created the Pari Center for New Learning.

Reviews

[Tom Schulte · Rating 3 out of 5]

It is interesting to me how wide in scope and populated the AI field was at the time for this 1985 book to survey. As the author reviews the stage of the art and work under way, it strikes me as a musing about human intelligence -- it is being modeled, imitated, and compared to -- and the pathways of input: the senses. Ersatz vision, hearing, speech, touch solutions are examined with "expert systems" and other thought-like software solutions. There is a nice summary in the final pages:

Yet despite the advances that have been made in expert systems, robotics, and problem solving the field of artificial intelligence is still in its infancy. As Daniel Dennett from Tufts University put it during a conversation with Jonathan Miller. "If you look at the actual products of artificial intelligence you find they're a relatively unimpressive lot; they're typically a bag of tricks and even when they do mimic a human being, its usually for spurious reasons. But one shouldn't judge the field by those gimmicks and illustrations, which is really what they are. The real products of the field are conceptual." Artificial intelligence, therefore, is a field full of challengingly difficult problems. By the end of this century it promises, along with the question of human consciousness, to be the most intellectually exciting area of research around.

In a sense this book is premature. It is as if a pre-Newtonian had attempted to map out the course of 20th-century physics with its quarks and black holes. The future of artificial intelligence is unpredictable and its advances are impossible to foretell. In a sense we are all waiting for its Einstein to emerge.