ual Machines, author Ray Kurzweil gives a very brief history of the Universe, which serves as a preface for his subsequent theories. In this history, Kurzweil chronicles the rapid expansion of time between salient events in the history of the Universe, describing time, in his own words, as geometrically slowing (pg. 10). He then jumps headfirst into the history of evolution, and shortly thereafter of technology, in both of which the time between salient events is shrinking exponentially. This leads him to question the opposing nature of the trend (how can time be accelerating as applied to technology and evolution yet decelerating as applied to the very Universe which contains these?) as well as search for similarities between the trends. Thus is created Kurzweils first theoretical law, that of time and chaos.
Kurzweils Law of Time and Chaos is as follows; “In a process, the time interval between salient events (that is, events that change the nature of the process, or significantly alter the future of the process) expands or contracts along with the amount of chaos.” (pg.29) In other words, as things become more chaotic as applied to a specific process it takes longer for significant events to occur within that process, and vice-versa. According to this law, the rate at which we advance technologically has, and will continue to, accelerate exponentially.
What if this exponential growth “hits a wall” so to speak, as trends of the exponential variety frequently do? Kurzweil is quick to answer this question, which he knows will be raised quickly by most readers. According to the Law of Accelerating Returns, which states simply that as a process speeds up so do the returns from that process speed up as well, technology will continue to build upon and advance itself. As technology advances, we are able to create more technologically advanced machines, which in turn will enable us to create even more advanced machines, and so on. According to Kurzweil, the only two resources this technological evolution needs to survive are “the growing order of the evolving technology…and the chaos from which an evolutionary process draws its options for further diversity” (pg 35), both of which, Kurzweil claims, are unbounded.
With the an understanding of the Laws of Accelerating Returns and of Time and Chaos firmly under our belts, Kurzweil advances to the next chapter in order to answer a question subtly raised by his faith in the continuing exponential advancement of technologycan an intelligence (such as ours) create an intelligence (such as the artificial intelligence of our computers) more intelligent than itself?
His answer is yes, and he comes to this conclusion by looking at the process of human evolution as an intelligence in itself. If this is the case, and we are to measure intelligence in terms of speed and frequency of error (as we do for an IQ test), then evolution, according to Kurzweil, would rate “only infinitesimally greater than zero” on that same IQ test. Therefore, humansa creation resulting from the intelligence of evolutionare more intelligent than the intelligence that spawned them. Kurzweil cites the example of scientists ongoing work with DNA, which is on the brink of allowing us to refine and control evolution as the original process never could, as evidence that we have indeed become more intelligent than the process that gave us birth.
It is not a difficult comparison which leads Kurzweil to postulate that some day computers, the intelligence that man created, will some day become more intelligent than man himself. It is also not difficult to foresee the day when computers more intelligent than man will begin to create intelligence more intelligent than theythus Kurzweil brings to a close his second chapter, setting nicely the stage for the rest of the book.