“Inventing Is a Lot Like Surfing”
RAYMOND KURZWEIL IS ONE OF THE MOST SUCCESSFUL INDEPENDENT INVENTORS LIVING-AND ONE OF THE MOST AUDACIOUS TECHNOLOGICAL PROPHETS.
RAYMOND KURZWEIL IS ONE OF THE VERY FEW CONTEMPO rary inventors who come close to the Thomas Edison loneinventor stereotype. A child prodigy who made numerous television appearances demonstrating his creations, he won first prize in electronics and communications at the International Science Fair as a teenager and paid for his college education by selling computer programs he wrote. The most amazing of his inventions so far may be the Kurzweil Reading Machine, introduced in 1976. Designed to give the visually impaired unfettered access to the world of books, it introduced two independent inventions: the CCD (charge-coupled device) flatbed scanner and the text-to-voice synthesizer. Over the years, Kurzweil and his partners have formed nine companies to develop products ranging from teaching machines and computerized portfolio management to the Kurzweil 250 music synthesizer.
He’s probably best known, though, for two books: The Age of Intelligent Machines , published in 1990, and The Age of Spiritual Machines , from 1998. They look at technological trends to predict startling changes in the decades to come, including the emergence of computers that can surpass the myriad abilities of the human brain. He has been the recipient of awards ranging from the National Medal of Technology to the Lemelson-MIT Prize. We talked in his office near Boston after he was named to the class of 2002 of the National Inventors Hall of Fame.
How did you decide to become an inventor?
From the time I was five I thought I would be a scientist, and to me that meant an inventor. To me a scientist was someone who took ideas and created things that would have an impact on the world. Not that my ideas were realistic. I tried to build a moon rocket. At eight I started having ideas with a little more traction. I built a robotic puppet theater that would bring in different characters and scenery automatically.
I was very much influenced by the Tom Swift, Jr., books. Tom Swift and his friends, and usually the whole human race, would get into some terrible jam. Tom and his friends would then go into their lair and come up with an invention that would save the day. The idea that human thought could come up with a solution to any problem really inspired me.
How did this lead to your work with computers?
I got involved with computers in 1960, when I was 12.1 obtained access to them, started building my own, and started programming. I had the use of punch-card machines. You prepared your programs in FORTRAN on punch cards and fed them in. The computers took up a whole room, and they were less powerful than the computer on your wristwatch today. When I came to MIT in 1965, all the students and professors shared one computer, an IBM 7094. It had a sixth of a megabyte of memory and was 2,000 times slower than today’s personal computers. But I became fascinated with the idea that you could model the world in a computer. One of my first projects was to model the creative process to produce melodies. I had a program that would analyze the patterns in famous melodies and then create original ones in a similar style.
How did you come to develop the reading machine?
Initially my interest was in a classic unsolved problem in pattern recognition: recognizing typed and printed letters in any font and dealing with all the vagaries of print. We put a team together and we created what’s called omnifont character recognition. This then became a solution in search of a problem. What was it good for?
One day on an airplane trip, when I happened to sit next to a blind gentleman, he told me, “There’s really no handicap associated with being blind.” Then he corrected himself. He said, “There is one problem: timely access to printed matter, like books and magazines.” Relatively few books ever get translated into Braille; material like office memos almost never does. He said that if he had a machine that let him take any printed matter and read it immediately, he’d overcome the principal handicap associated with blindness. We put it all together, and we introduced it in 1976.
Were you concerned with making a profit?
There’s not necessarily any conflict between doing something culturally significant and something successful. Technologies of the reading machine spawned other technologies—scanners, speech synthesis, optical character recognition—that have had many significant applications.
Did you foresee all these applications?
No. We especially didn’t appreciate how helpful this would be in making databases that span the world over the World Wide Web, by scanning material from print and understanding it through character recognition and data-mining tools. We did have the idea of offering the reader to people who had reading problems for reasons other than blindness—kids with reading disabilities, for example—but we didn’t realize how much it would help them by acting like a reading teacher. A dyslexic child can’t see the print until he’s exposed to the correlation between it and sound. The reading machine does essentially what a teacher would do: read to the child. And it can highlight what is being read at the time.
Have you had good luck protecting your inventions?
Yes, with a combination of patents, copyrights, and trade secrets. It’s certainly critical. The power of technology has created new challenges to intellectual-property protection, since you can so easily copy movies, music, and software, but you need that protection or you can’t create the business model that will provide the financing to create the intellectual property in the first place.
Do you enjoy running the business as much as inventing?
I have an interest, because the practical business side of technology is part of the process of inventing. What turns me on is the linkage between a dry formula on a blackboard and a change in people’s lives, the series of steps that come in between. Part of that is creating a business. What excites me is the power of an idea to change the world, and if you’ve got a cool project that sits on the table and makes a nice demonstration, it doesn’t change the world. You’ve got to get it out there.
Timing is also very important. Inventing is a lot like surfing. You’ve got to catch the wave. The world will be a different place by the time you’ve completed your project, and most projects fail not because things don’t work but because the timing is wrong. The technology or market isn’t really ready, or it’s too late and you’ve missed the window. This is what has made me an ardent student of how technology evolves. What will computers be like, what will the channels of distribution be like, three or five years from now?
Which points toward your books. How did you decide to start writing?
My interest in writing goes way back. I actually majored in both computer science and creative writing at MIT, and I studied with Lillian Hellman. Writing is another way of altering the world, of inventing. We need to build bridges between the science and engineering communities and the rest of the population. They say war is too important to leave to the generals. Well, technology is too important to leave to the scientists and engineers. We need the whole society to become literate about what is feasible in technology so we can intelligently address its promises and its perils, which I believe are deeply intertwined.
I have a whole team of people gathering data about technology trends, and I’ve been developing mathematical models of how these trends evolve. I have a theory that the evolution of technology is an extension of the evolutionary process that created our species in the first place. I believe technology evolves according to its own rules and methods, and I’ve tried to share my vision of how the future will be different from today.
People today realize that there is change, that the only constant is change, but they don’t realize how the pace of change is changing. We’re actually doubling what I call the paradigm shift rate every 10 years. We’ll make 100 years of progress in the next 25 years. It’s exponential growth, and it’s in its nature that it can explode. We’re at that point now with a lot of our technologies. Computers and communications technologies are getting enormously powerful. By combining them with more powerful software, improved by our insight into how human intelligence works, we’ll get exponentially growing knowledge about the human brain.
How have readers reacted to your predictions?
There has been a sea change. When my last book came out, four years ago, it was considered radical. Today it’s more mainstream, as the educated public has been bombarded with so many announcements of breakthroughs in various fields. What I’ve written seems more reasonable to people.
Can you really predict the future with any precision?
Nobody has a perfect crystal ball, and there certainly are unpredictable events, but what impresses me is just how predictable some trends are, despite major dislocations. Even major events of the twentieth century such as World War II and the Depression didn’t really affect the trends. They just kept marching along. E-commerce has progressed very smoothly through both boom and bust. Telecommunications has too, if you look at it in terms of actual bandwidth and other objective measurements.
I’ve tried to bring some structured methodology to understanding technology trends and predicting the future. Most prognoses are done very informally, without any methodology. I’ve just tried to bring some precision to that.
