by Flemming Funch
Via FuturePositive, Ray Kurzweil being interviewed on the accelerating rate of change.The Law of Accelerating Returns is the acceleration of technology, and the evolutionary growth of the products of an evolutionary process. And this really goes back to the roots of biological evolution.
Evolution works through indirection. You create something and then work through that to create the next stage. And for that reason, the next stage is more powerful, and happens more quickly. And that has been accelerating ever since the dawn of evolution on this planet.
The first stage of evolution took billions of years. DNA was being created and that was very significant because it was like a little computer, and an information processing method to store the results of experiments, and to build up a knowledge base from which it could then launch experiments and codify the results.
The subsequent stages of evolution happened much more quickly. The Cambrian Explosion only took a few tens of millions of years to establish the body plan to evolve animals. And we see that evolution, like certain technologies, has become mature and stopped evolving. Evolution has concentrated on other issues, specifically higher cortical functions. And that happened much more quickly than the Cambrian Explosion. Humanoids evolved over many millions of years, and Homo sapiens over only hundreds of thousands of years. And there again, evolution used the products of its evolutionary processes, which was Homo sapiens, to create the next stage, which was human-directed technology, which really is a continuation of the cutting-edge of the evolutionary process on earth, for creating more intelligent systems.
In the first stage of human-directed technology, it took tens of thousands of years, which is what you would expect for the next stage via the wheel, or stone tools, and that kept accelerating, because when we had stone tools, we could use them to build the next stage. So a thousand years ago a paradigm shift only took a century, like the printing press. And now a paradigm shift, like the World Wide Web, is measured in only a few years’ time. The first computers were built with screwdrivers and were designed with pencil and paper, and today we use computers to create computers. A CAD designer will sit down and specify a few high-level parameters, and 12 different layers of automated designs will be done automatically. The most significant acceleration is in the paradigm shift rate itself, which I think of as the rate of technical progress. And all of these are actually not exponential, but double exponentials because not only does the process accelerate because of our evolution’s ability to use each stage of evolution to build the next stage, but also, as the process, as an area gets higher price performance, more resources get drawn into that capability.[..]
The whole 20th century, because we’ve been speeding up to this point, is equivalent to 20 years of progress at today’s rate of progress, and we’ll make another 20 years of progress at today’s rate of progress equal to the whole 20th century in the next 14 years, and then we’ll do it again in seven years. And because of the explosive power of exponential growth, the 21st century will be equivalent to 20,000 years of progress at today’s rate of progress, which is a thousand times greater than the 20th century, which was no slouch to change. Kurzweil is one of the proponents of The Singularity - the idea that a number of accelerating technological trends are going to converge in a way that will totally transform our existence. In our lifetimes. Nanotechnology, Artificial Intelligence, Genetic Engineering, and more. Personally, I agree that there's something like that going on, and that life as we know it will totally change, but I don't see it quite as materialistically. I think WE are evolving and transforming WITH and THROUGH technology. Which is a very risky thing to do so quickly. But I don't quite go along with the idea that one of our main concerns will be that robots will become smarter than us.
|
|