|
94 of 103 people found the following review helpful:
3.0 out of 5 stars
Lacking depth sometimes, but still very interesting., September 7, 2000
Ray Kurzweil's "The Age of Spiritual Machines' is an intelligent look at what the future might be holding for us all. Like other similar titles - Visions by Michio Kaku comes to mind - Kurzweil tries to predict where science will take us. Unlike `Visions' however, this book is considerably more focused on computer technology and artificial intelligence, and I would only recommend it if you're not looking for a much broader answer to the question of where we are headed. Kurzweil never intended to cover other matters, and reading the Prologue will be enough to understand that most of the book will explore the rising of machine intelligence to a level that will surpass the capabilities of the human brain. Kurzweil starts by describing the exponential growth of computer power, Moore's Law, and transistor-based computing. The present and the future are described until quantum effects start becoming a problem and a completely new kind of technology becomes necessary (some alternatives are mentioned, Quantum computation is of course, mentioned). The book proceeds to more metaphysical subjects, and questions if we can create another intelligence form more intelligent than ourselves. Can the created exceed the creator? It will then proceed to cover consciousness and feelings; Kurzweil gets philosophical in what in my opinion is one of the book's weakest chapters The methods available to solve a wide range of intelligent problems (when combined with heavy doses of computation) will follow, in a chapter that covers subjects from recursive formulas to neural nets, and of course, enough space is dedicated to Alan Turing, the father of all modern computers. Part 2 starts with my favourite chapter of this title; Kurzweil discusses how evolution has found a way around the computational limitations of normal neural circuitry. And from nature's lessons we move to ideas about molecular computing harnessing the DNA molecule itself as a practical computing device, now a possibility under investigation. I wish I had this book last year when I was doing some research on general quantum computing for college, Kurzweil fully managed to transmit the impact that future developments in these areas might cause, and the problems that will be caused by ultra-fast parallel computation (especially with cryptography). The port of slow mammalian carbon-based neurons to speedier electronic and photonic equivalents is covered with simplicity, but convincingly. Next comes the problem of the body. A disembodied mind will quickly get depressed, no matter how powerful. So what kind of bodies should our machines have, or later on, what kind of bodies will they provide for themselves? Part 2 ends with a few thoughts on the array of tasks that are now performed by computers, lacking sense of humour, talent for small talk and other endearing qualities, but still vital for tasks that previously required human intelligence: How much do we depend on modern technology? If all the computers stopped functioning, would chaos rise? Is our world too based on technology and vulnerable to global disasters? After 2009, the book truly starts facing the future. You will be shown how extremely cheap and powerful (compared to today's standards) computers will be imbedded in clothing and jewellery, among other items, surrounding us completely. Virtual personalities start emerging, and Kurzweil dares to predict real time translating telephones and even human musicians jamming routinely with cybernetic musicians. Also interesting, I thought, is the possibility of some sort of neo-Luddite movement growing around this time. Next stage is 2019. By this time, Kurzweil believes that a $1k computing device will be approximately equal to the computational ability of the human brain. Computers should be almost invisible, and will be everywhere. 3D virtual reality will reach good quality levels, and VR displays are embedded in glasses and contacts lenses, providing a new interface (and the main interface) for communication with other persons (via the future version of the Web). Interaction with computers is made through gestures and 2-way natural language. A few thoughts on relationships with automated personalities end the chapter. By 2029, Kurzweil's predictions turn to direct neural pathways that somehow have perfected some soft of high-bandwidth connection to the human brain. Ultra fast learning à-la-Neo from Matrix in less than 28 years? Kurzweil suspects so. Neural implants become widely available to enhance visual and auditory perception and interpretation, as well as memory and reasoning. People with physical problems and strongly helped by implants. Computers have "read" all available human literature and the discussion about legal rights of computers and what constitutes being human. Machines claim to be conscious. Around 2099, human thinking starts merging with the world of machine intelligence. There is no clear distinction between humans and computers. Most of the intelligences are not tied to a specific processing unit, but widely spread. This chapter's most interesting aspect is perhaps the discussion about software based humans, when compared to those still using carbon-based neurons. The use of neural implant technology provides enormous augmentation of human perceptual and cognitive abilities, creating some sort of division between first class and second-class humans. Kurzweil implies that those who do not utilize such "enhancements" will be unable to meaningfully participate in dialogues with those who do. Being alive no longer means what it used to mean. Life expectancy is no longer a viable term in relation to intelligent, machine-based intelligent beings. The books ends with a few thoughts on the fate of the whole universe, a part that is probably the weakest of the whole book, extremely pale when compared to Michio Kaku's "Visions" look. Kurzweil might do a good job describing a universe in which artificial intelligence and nanotechnology combine to bring longevity, but failed partially when discussing that longevity and the coming connections of computers with immortality, a subject that deserved a lot more attention and space in this book. Left me wanting more. You will find this book fascinating if you're particularly interested in what the future holds when it comes to computers. Kurzweil knows his science well and adding a bit of common sense and humour, is enough to result in a very enjoyable title. If the predictions turn to follow the expected timeline, well, frankly I don't care much, and I don't think it's very relevant to discuss it; Most of it will happen precisely as the author puts it, but it might take more or less time. This book is not complex, and has many references and notes; so even people with a poor background in computer science will be able to follow the author's ideas. Of course knowing what's behind it will make your experience a lot richer. You also get a decent glossary, very valuable if you're new to the subject. Overall, a good book, but lacking depth in some areas (especially machine based existence and immortality). Sometimes too over simplified. Still, check it out and see where we're heading. Combine it with Neuromancer, Visions and a few more technical titles and you will wish you could live 300 years...then again, maybe not. ;-)
|
by Ray Kurzweil
by Joel Garreau
by Ray Kurzweil Ph.D.
by Ray Kurzweil
See all 100 books this book cites
