HBO recently adapted the first book in the four-part series for the small screen.
Inventor, futurist and author Ray Kurzweil has long predicted humans will one day be able to transcend the limitations of their biology. In a new book, Kurzweil explains why that day is coming sooner than we might think. He argues that the expansion of the brain’s neocortex was the last biological evolution man needed to make. That’s because it is inevitably leading to “truly intelligent machines,” which Kurzweil calls the last invention that humanity needs to make. Join Diane and Ray Kurzweil for a discussion on prospects for attaining immortality through technology.
- Ray Kurzweil Inventor, futurist and author.
Photo Gallery: Computer Simulations Of The Nervous System
Reprinted by arrangement with Viking, a member of Penguin Group (USA) Inc., from “How to Create a Mind.” Copyright © Ray Kurzweil, 2012.
In this 2005 TED talk, inventor, entrepreneur and visionary Ray Kurzweil explains in abundant, grounded detail why, by the 2020s, we will have reverse-engineered the human brain and nanobots will be operating your consciousness.
Read An Excerpt
Reprinted by arrangement with Viking, a member of Penguin Group (USA) Inc., from “How to Create a Mind.” Copyright © Ray Kurzweil, 2012.
MS. DIANE REHMThanks for joining us. I'm Diane Rehm. Ray Kurzweil could be described as a techno-evangelist. His belief in the power and good of technology has almost a religious fervor to it, but it's grounded in science.
MS. DIANE REHMIn a new book, he takes the reader on an exploration of the evolution and wonders of the human brain. He explains how scientists are attempting to reverse-engineer the brain and why success could lead to immortality.
MS. DIANE REHMThe book is titled "How to Create a Mind: The Secret of Human Thought Revealed." Author Ray Kurzweil joins me in the studio. You're welcome to be a part of the program. Call us on 800-433-8850. Send us your email to firstname.lastname@example.org. Follow us on Facebook or Twitter. Welcome, Ray, it's good to see you again.
MR. RAY KURZWEILIt's great to be with you again, Diane.
REHMThank you, Ray. Forbes Magazine calls you the ultimate thinking machine. Early in your book, you say predicting the future is actually the primary reason we have a brain. Explain.
KURZWEILWell, that's a good citation from the book because that's why we have a brain, to anticipate the impact of our decisions and lack of decisions. So when we walked through the savannah 1,000 years ago and I saw an animal walking towards a rock and I could see myself walking towards the same rock, I would make a prediction that we're going to meet.
KURZWEILAnd I would then make a decision, no, I'm going to take a different path. And that turned out to be good for survival. That's why we have a brain, in order to be able to anticipate the future.
KURZWEILThere's a problem, though, with our predictors in the brain in that they're linear. We assume that animal is going to keep going the same speed. That works well for animals. It's not actually good for predicting information technology.
REHMSo how is predicting information technology very different? Not linear, I presume?
KURZWEILRight, it's not linear. It's exponential. It speeds up. For example, the power of computers is doubling every year and so it goes 2, 4, 8, 16 rather than 1, 2, 3, 4. Now that doesn't sound so different, but by the time you get to the 30th step of a linear progression, that's our intuition getting you to 30.
KURZWEILThe reality, not of everything, but of information technology is if you double 30 times, you're up to a billion and that's not an idle speculation about the future. The cell phone I have on my belt is a billion times more powerful for the same cost as the computer I used when I was a student.
REHMI want to make sure that cell phone you have on your belt is turned off.
KURZWEILYes, it is off.
KURZWEILSo it's not very powerful at the moment.
REHMIf thinking about the future is not linear, what about creating memories, somehow pulling and holding?
KURZWEILWell, this is what the book is about, which is we have a region of the brain called the neocortex. It emerged in mammals and it's gotten bigger and bigger as mammals have evolved.
REHMWhere is it?
KURZWEILIt's on the outside of the brain. Neocortex means new rind so it covers the brain. It's about the size of a table napkin in a human. It's about the thickness of a table napkin, but it has many folds and convolutions,, if you remember the picture of a brain.
KURZWEILSo it's actually now 80 percent of the weight of the brain and we actually have a little bit more than other primates because we have this big forehead with the frontal cortex that enables us to think higher thoughts. That was the enabling feature of the brain that permitted the evolution of language and art and science.
KURZWEILWe basically have a bigger neocortex and what's unique about the neocortex is that it enables us to think in hierarchies. We're not just thinking of one thing after another. We have this elaborate hierarchy. If you think of language and concepts, something like irony or beauty, it's a very complex concept that's made up of other concepts, which are made up of other concepts.
KURZWEILWe can take a whole bunch of ideas, call that a new idea, give that a symbol and then use that symbol with yet other ideas and create another symbol and build up this whole hierarchy we call knowledge. Only the neocortex is able to do that. And it enables us to solve problems and learn new skills that are flexible.
KURZWEILIf you don't have a neocortex, such as animals before mammals, you can do rote learning. You can learn a certain skill that doesn't have much flexibility.
REHMSo you're citing, for example, the difference between humans and dogs?
KURZWEILWell, dogs are mammals and they have a neocortex and so they are able to learn skills that have some flexibility. They can. Say a mouse is trying to escape from a cat. It can do that even if the path is blocked. It can actually figure out a new route. So it has a certain amount of this type of intelligence.
KURZWEILIts neocortex is obviously a lot smaller than ours. Animals like lizards don't have any neocortex at all. Their behavior is pre-programmed. So when this catastrophic event happened 65 million years ago called the Cretaceous Extinction Event, probably caused by a meteor and the environment changed, suddenly the non-mammalian species were not able to adapt and tens of thousands of those species died out. And that's when mammals kind of took over their archeological niche.
KURZWEILAnd from that point on, the neocortex kept growing because to anthropomorphize the neocortex, the biological evolution figured, wow, this is pretty good stuff, this neocortex, and it kept growing it as mammals evolved.
REHMAnd is continuing to grow?
KURZWEILWell, there actually are genetic changes and biological evolution has continued even in the last few thousand years, but that's really insignificant compared to technological evolution. The real continuation of evolution now is in our technology and that's moving a million times faster than biological evolution.
KURZWEILSo what my book is about is to actually understand, use our human intelligence to understand how human intelligence works and then use that to create systems that are similar, based on the same principles and then use it to make ourselves smarter. And I would say we're doing that already. We've taken the first few steps, if you think about how, you know, Google and Wikipedia and all these other tools we have, these mind expanders are making us smarter already.
REHMSo let me understand. You're saying that the technology that we have developed is leading us on a path to become smarter and smarter. And what you're suggesting in the book is that we may become smart enough to become immortal?
KURZWEILRight. Well, that's jumping to the final...
REHMI'm really jumping.
KURZWEIL...thought. But the first observation is intelligence allows us to solve problems and it really requires the neocortex to solve problems in a creative fashion. Our neocortex can create metaphors so Darwin noticed a metaphor from geology that tiny, little trickles of water can carve out a great cavern.
KURZWEILHe said, ah, that's just like the tiny trickle of changes from one generation to another that can carve out a new species. So he made a metaphor from geology to biology and that's how science evolves. Literature, art is full of metaphors and that's what the neocortex is good at.
KURZWEILAnd the neocortex has already enabled us to overcome many problems. Human life expectancy was 20 a thousand years ago. It was 37 in 1800. We are wealthier, healthier than we were hundreds of years ago because of our intelligence. We're now actually augmenting our intelligence.
KURZWEILWe've expanded our memory, for example. You know, you can, at the dinner table, find out the answer to at least factual questions in a few seconds. We have at some level, all of human knowledge at our fingertips. That's because we have codified our knowledge and created technology that can encompass some of it.
KURZWEILThis is the field of artificial intelligence. That's been my interest for 50 years. It's really evolved on its own without an understanding of how the best example we have of human works, which is the human brain, because we have not been able to actually see inside the brain with enough precision up until just recently.
KURZWEILNow we can actually see with enough precision to ascertain how, you know, what are the methods that are used in the brain? And then we can use these as inspiration, biologically-inspired paradigms to create intelligent machines that work the same way. And if we look at some recent examples like Watson, the IBM computer that won the "Jeopardy" game on television and actually got a higher score on "Jeopardy" than the best two human players put together.
KURZWEILAnd that's not a narrow game. It involves understanding human language and understanding puns and metaphors and similes and jokes. And then it has all of human knowledge, at least some level at its metaphorical fingertips.
KURZWEILIt got its knowledge by reading Wikipedia. So it was not hand-coated by the scientists. So it has some level of being actually able to understand, you know, normal human language. That's one, you know, impressive example of how far we've come already.
REHMBut, you know, if you're sitting around the Thanksgiving table and you want to know something, a fact, you turn to a machine to find out the answer. Does that actually expand knowledge or does it simply implant a fact?
KURZWEILWell, it has, you know, it's limited today. We can find certain specific items of knowledge either from our own memories or from our collective memories so we've kind of outsourced portions of our memory to our machines. These systems, though, are going to get more and more intelligent.
KURZWEILSearch engines in a few years won't just be based on looking for words, they'll actually be looking for concepts and they'll be able to read these billions of web pages and understand what they're talking about.
REHMRay Kurzweil, inventor, futurist, author of numerous books, his latest is titled "How to Create a Mind." Stay with us.
REHMAnd welcome back. I'm going to read you this first email because it's such a good one. It's from Dean in Rockville, Md. who said, "In his book "Singularity" Kurzweil predicted artificial intelligence would surpass human intelligence and perhaps merge with it by the year 2020. Does he still believe in that schedule?
KURZWEILWell, actually my consistent prediction has been 2029. Computers will pass something called the (word?) test, which basically means they'll be inextinguishable from human intelligence. But they then combine that with the speed and scale and total recall that computers have. So consider Watson, which I mentioned a few minutes ago. It's read all of Wikipedia. It actually doesn't understand it as well as you or I, however it could read all of it. It read 200 million pages of natural language documents. It can recall any of that in three seconds even though its understanding is not at the human level.
KURZWEILSo imagine it actually could read at human levels and then read everything on the web and really understand its meaning. That's where we're headed. But again, this is not an alien invasion as our writer -- caller points out. We're going to merge with these technologies. That's why we create these tools. We've -- ever since we've picked up a branch and fashioned it into a tool to reach a higher branch we've been extending our reach physically and now mentally with our tools. That's what technology is all about.
KURZWEILAnd we're going to literally make ourselves smarter. We're at the beginning steps of that already. You know, during that one day SOPA strike I felt like a part of my brain went on strike when we couldn't access, you know, brain extenders like Wikipedia and Google. So it's already part of who we are and that's why we create these tools. We will make ourselves smarter. There's limitations to the neocortex.
KURZWEILWe only have, for example, 300 million pattern recognizers. We can recognize 300 million patterns. That sounds like a lot. It was a lot. It enabled us to create art and science and technology. But it's also a limitation. We'll be able to expand the scope and scale of our own thinking by emerging with these brain extenders. So 2029 is the date...
REHM2029, not 2020.
KURZWEILRight. That's actually a time where the hardware, the speed and capacity of computing will be sufficient, but we won't have the software which is -- which are the methods. And that's why my book is about, describing what the methods are in the human brain and how we can harness that to create even more intelligent machines.
REHMAnd before we even get to that I'd like to know who Ray Kurzweil is and how...
KURZWEILI was hoping you would tell me.
REHMYeah, and how you got to be who you are? What were your influences in childhood?
KURZWEILWell, my parents liked to give me all these enrichment toys, things like erector sets that had lots of little pieces in them. And at five years old I had this big inventory because I took apart all these different toys and had lots of different parts. And I went through the neighborhood and brought back broken radios and broken bicycles and took them apart. This was an area where you would allow a five-year-old to do that. And I had this idea if I could just figure out how to put these things together in the right way I could create these sort of magical effects. I could solve any problem.
KURZWEILWhat I realized a few years later, as I really got this idea from my family because I was kind of the family religion, the power of human ideas. I remember my grandfather coming back when I was seven from his first return trip to Europe after having fled Vienna in 1938 from the Nazis. And he described with almost reverential terms the opportunity to handle with his own hands some original documents by Leonardo De Vinci.
KURZWEILSo these were sacred documents, but they weren't handed down by god. They were created by a human and they were able to change the world because of the power of these ideas. So that was kind of the family philosophy and I fashioned myself an inventor from when I was five. All the kids were wondering what they wanted to be and I had this conceived, well I know what I'm going to be. And it took a while for my inventions to get any traction but I did have that idea from an early age.
REHMAnd what were you like as a school child?
KURZWEILI was working on my projects and I still am. One project led to another. I had these sort of virtual reality world I created that did kind of work when I was eight using mechanical linkages. I could move the sun, stars and trees on or off the stage from my command station. I discovered the computer at 12. That's not so amazing today. It's hard to find a 12-year-old who doesn't use the computer. But there were only maybe a dozen computers in all of New York City when I was 12 in 1960.
KURZWEILAnd I did have the idea then that we could really -- we could recreate the world's virtual realities and ultimately our thinking. I wrote a paper when I was 14, which is my Westinghouse Science Talent Search submissions, not the Intel Science Talent Search. And I got to meet President Johnson. I was one of the national winners. Created a system that could recognize patterns and musical melodies and then write original melodies in the same style. I took it on the TV program "I've Got a Secret," which most of your listeners probably will recognize.
REHMI remember that. I remember that.
KURZWEILAnd what the paper said is that the heart of human intelligence is recognizing patterns -- pattern recognition. That is what the brain does really well. We're not very good at logical thinking. Even back then with the primitive computers we had in 1960 computers were better at logical thinking but we could recognize patterns. And that is actually the message of this new book 50 years later. But I describe in real detail how the neocortex does that. We have basically 300 million little regions that recognize patterns and they're organizing hierarchies. So I'll have one little pattern recognizer that fires when it--when it sees a crossbar in a capital A. And it's all it cares about. It gets very excited when it sees a crossbar in a capital A.
KURZWEILAt a higher level there'll be a recognizer that recognizes a capital A. At a yet higher level there'll be a recognizer that goes ah, it's the word apple. Go up another 20 levels there'll be a recognizer that's getting input from different senses. It sees a certain fabric, smells a certain perfume, hears a certain voice and goes ah-ha, my wife has entered the room. Go up another 20, 30 levels as a recognizer goes, that's funny, she's pretty. That was ironic. You might think that those higher level more abstract recognizers are more complicated, more sophisticated.
KURZWEILThey're actually basically the same except for their position in this grand hierarchy. And that's the essence of the neocortex. It's organized in this very elaborate hierarchy from very low level primitive -- recognizes -- it just recognizes the edges of objects and simple things up to these very abstract features. And the key is that the -- we're not born with the hierarchy. We create that from our own experience. Everybody's hierarchy is different because it's not only are you what you eat but you are what you think. With every thought we make we create that elaborate hierarchy.
REHMSo you're suggesting that thinking in certain ways can indeed enhance our thoughts.
KURZWEILAbsolutely. I mean, you are what you think so be careful who you hang out with. I think listening to "The Diane Rehm Show" is a good bet. But literally from the moment you're born or even earlier -- because our eyes open at age 26 weeks -- we are laying down one conceptual level at a time. That's another key is that we can only sort of learn one level of abstraction at a time. So I've got a one-year-old grandson now and so he's managed to lay down a few levels. But he's, you know, still learning the primitive features and patterns and language and so on.
REHMAnd how are you interacting with him in order to help him expand the development of that neocortex?
KURZWEILWell, I'll -- advice is still good advice to expose children at any age to a very rich and varied environment. And reading him stories, which he doesn't quite understand conceptually but can -- is delighted at the sound of the words and the variation and tone and emotion. And so exposing children to a broad variety of ideas that are age appropriate is the right approach because they are building that neocortex, which will define their personality, the skills, their ideas, their memories, their thoughts.
REHMSo at your own age now, 65, 66, how are you continuing to develop your neocortex?
KURZWEILWell, that's a great question because we actually fill up these 300 million pattern recognizers by the time we're 20. And in order to learn something new at that point we need to forget something old. That's one reason why say a five-year-old can learn a new language so easily. They've got all this virgin neocortex. They can just learn something new and fill it up with new thoughts.
KURZWEILNow there's a lot of redundancy. I mentioned I have pattern recognizers in my neocortex that recognizes a crossbar in a capital A. I don't just have one. I've got probably thousands of those and I can give up some because, you know, probably half as many would be just as good.
REHMConsciously give them up?
KURZWEILIt's not a conscious process, but it's a necessary process. We have to forget something including some redundancy. So that's why in fact an old memory will dim because, you know, originally we may have it hundreds of times. And if we, you know, are using that neocortex for other ideas and thoughts we'll give up a lot of that repetition and the memory will become more dim over time. Now some people are better at giving up old ideas than others. I mean, many people are rigid thinkers and just happy with the neocortex they have and don't want to entertain new ideas.
REHMI was about to say, you have to want to move forward.
KURZWEILI think that's right. I mean, intention is very important and the idea of actually being delighted in new ideas. So that was a philosophy my family had is to always be open to new ideas. And that means being willing to give up old ideas.
REHMHow -- give me an example of a new idea you would wish to entertain and absorb and an example of an old idea you'd give up in order to do that.
KURZWEILWell, there are many orthodoxies saying in every field -- and I talk about them actually in the book the orthodoxies that existed in biology before Darwin or in physics before Einstein. And people were very happy with those structures and those concepts, even though they began to contradict reality. And, you know, those thinkers got a lot of resistance as a result. So it's very important to actually pay attention to new evidence and new ideas.
KURZWEILAnd our ability to give up old ideas, I don't have a magic formula for how we do that. It is a subconscious or unconscious process. Some people are better at it than others. Most people are not very good at it because the vast majority of say scientists at the time of those legendary figures were very resistant to a new way of thinking.
REHMWanting to hold on to what they thought they knew.
KURZWEILIt's comforting to have explanations for everything. We have to realize the limitations of our ideas to gain new ones.
REHMAnd you're listening to "The Diane Rehm Show." Here's another email. "Why is it that the brain seems to be so much more powerful than evolutionary experience would suggest?"
KURZWEILWell, there's two ways to learn things. These animals that did not have a neocortex, basically non-mammalian species, can learn new things too but just not in the course of one lifetime. So they have preprogrammed behaviors, go back a hundred million years ago when they could learn something new over say a thousand generations to the process of biological evolutions, which Darwin has explained. And so they would create a new behavior, you know, over 20,000 years.
KURZWEILThat was actually good enough 100 million years ago because the environment changed so slowly they could keep up with it through this process of evolution. It wasn't good enough 65 million years ago when we had this sudden catastrophic event, probably caused by a meteor and the environment changed very suddenly. And so many of these species without a neocortex died out and that's when mammals took over their ecological niche. Because they had a neocortex and therefore had a more flexible way of thinking. They could learn new skills. They could adapt them flexibly based on the natural hierarchy.
KURZWEILThe world his hierarchical and you really can't understand the world unless you can understand its natural hierarchy. The fact that, you know, branches lead to other branches and the whole assembly of branches makes a tree and this assembly of trees make a forest. And this is a natural hierarchy to the world. You can't really understand it otherwise. This was the first mental organ that enabled animals to do that. And so we can -- the neocortex can solve problems in days potentially. And then it could sit and then that idea can spread, you know, throughout a species quickly by one member watching another and learning these hierarchical concepts.
KURZWEILWe have more neocortex than other mammals so we can -- that was the enabling factor to enable us to develop language and all of our technology and ideas, art and science and so on.
REHMAll right. I'm going to take a caller in Hanover, N.H. Good morning, Dave. You're on the air.
DAVEI'm just so thrilled to talk to you today.
DAVEI have a question. I'm a forest ecologist and I work with global change models. And my question is, we're using computers that are becoming rapidly more advanced and more (unintelligible) kind of the idea that computers will continue to build exponentially. What I'm curious about is we have the advanced computers that can predict -- that we use to predict into the future to answer questions. But evolutionarily our brain has been designed to only think one lifetime ahead based on our kind of evolutionary strategy.
DAVESo the question is, you know, we have all this mounting evidence for things like climate change, which is something that I study from these really advanced computers. If the decision-makers in our society can only think so far ahead because we're limited with our own brain and that too determines what happens, what good is that technology? I mean, how are we ever going to solve the discontinuity between our advanced predictive machines and our evolutionarily limited brain?
REHMAll right. Quick answer and then we'll come back to more of it.
KURZWEILWell, that's why we have technology. This exponential growth they call the law of accelerating returns, Moore's law is just one example of that. And we're able to see the real implications. You know, you're right that our brains are -- were shaped a thousand years ago and our problems were very different.
REHMAll right. And we'll complete that answer when we come back.
REHMAnd just before the break, Ray Kurzweil was responding to Dave's question about the speed with which climate change computer scientists see what's happening, and yet politically we're pretty slow.
KURZWEILWell, our unassisted brains are limited, that's why we create technology. And one example of technology are models where we can look into the future. It's important actually to model many different phenomenon. We see climate change and also pollution from use of fossil fuels. But there's also new technologies coming. Larry Page of Google and I did a study for the National Academy of Engineering on new energy sources and noted that solar energy, for example, is doubling.
KURZWEILIt's on an exponential rise, doubling every two years. It's now about seven doublings from meeting 100 percent of the world's energy needs and we have 10,000 times more sunlight than we need. I'm quite convinced that within 20 years, we'll have a radical change in our use of energy and will be moving away from fossil fuels. And by the way, I've talked to people in the sovereign countries like the Arab oil-producing countries and they believe that there'll be a radical difference in the value of fossil fuels within 20 years.
KURZWEILBut our ability to think of any of that, of the implications of climate change or the solutions is based on our ability to create technology and, you know, modeling these phenomena's one type of technology. That's why we created technology because our unassisted brain, which evolved, you know, tens of thousands of years ago, was optimized for solving the problems that primitive and woman encountered, like avoiding a predator in the wild.
KURZWEILAnd that's a different type of skill. But we do have now a technology that can anticipate the future with more powerful tools.
REHMDoes that mean you believe we will be able to ameliorate climate change?
KURZWEILYes. I think that the same technologies that are accelerating exponentially and making more and more powerful cell phones, for example, are also going to transform the world of resources. There was a very good book written by Peter Diamandis, my co-founder of Singularity University with me, called "Abundance" where he talked about how this exponentially growing information technologies will transform availability of energy, water, food, using new 21st century technologies, all of which are progressing in an exponential pace.
KURZWEILYou need to look at all of these things before declaring that we're going to run out of resources. We'd only run out of resources if we stick to 19th century technologies.
REHMHere's a tweet: Can Mr. Kurzweil say something about consciousness, a definition perhaps?
KURZWEILIt's a -- I have a chapter on these philosophical issues, consciousness, free will, identity. Consciousness has been debated since Plato's time. I believe it is a philosophical issue. There's no machine you could describe where you slide an entity and then a green light goes on. Okay, this one's conscious. No, this one's not. That doesn't have some philosophical assumptions built into it. And different philosophers would have different assumptions.
KURZWEILJohn (word?) would want to make sure it's squirting biological neurotransmitters, and otherwise the green light wouldn't go on. And Dan (word?) would want to make sure that it had a model of its own thinking, but it wouldn't really care what it's made out of. And it comes down to the conceptual difference between objectivity, which is science, and subjectivity, which is consciousness. It's really only immediately apparent to its possessor.
KURZWEILSo sometimes he'd say, well, it's a philosophical issue, there's no scientific falsifiable experiment you can design to absolutely tell you if something else is conscious. Therefore, it's an illusion and let's not waste time with it. That would be a mistake, in my view, because our whole moral system is based on consciousness. And our legal system is loosely based on our moral system. If I caused pain and suffering to someone or extinguished a consciousness.
KURZWEILYou know, those are crimes -- if I destroy property, it might be a crime. If it's causing someone suffering, it all has to with consciousness. So it really is a practical issue. We really have to determine who and what is conscious. And we don't agree on it. That underlies a lot of the debate, say, in animal rights or even the issue of abortion. When do you have a conscious person? So I have a chapter on this that examines the different points of view and actually has my own -- what I say is that you have to have a leap of faith.
KURZWEILPeople say, well, I'm a scientist, I don't have a leap of faith. But in fact, you have to have a leap of faith to get up in the morning. You have to know who and what is conscious in order to make real decisions about your actions. And I state my leap of faith, which is basically if something or some entity really is convincingly acts like it's conscious, I will accept that is a conscious being. So, for example, a character in a video game today does not really seem conscious. It says, I'm angry at you, but it doesn't have all the subtle cues that we really associate with that emotion.
REHMSo where does that leave you on the issue of abortion?
KURZWEILIt leaves me that when you have a conscious being, you have a person. So, you know, to look at the extremes, I don't think either conception or birth are the right points. Two days after conception, you do not have a conscious being. You have a clump of cells. One week before term, you do have a conscious being. It's really insignificantly different from a born child. In fact, it could be born and survive. You know, it's a slippery slope. Nobody likes to be on that slippery slope, so people take extreme positions on it.
REHMHere's an email which begins: Sir, do you not see the negative ramifications to increasing the knowledge of machines? Wouldn't their knowledge with the lack of consciousness would just create a powerful sociopath? And P.S., haven't you seen "The Terminator"?
KURZWEILI've seen a few of them and I've actually written quite a lot on what I call the promise versus peril of these technologies. And technology has been a double-edged sword ever since fire, which kept us warm, cooked our food, but also burn down our villages. And take biotechnology today, we're using it to reprogram biology away from disease, but could also be used by a bioterrorist to create a new super weapon.
KURZWEILWe actually have now a rapid response system which I was involved in advising the army on to counteract that, somebody actually put out, you know, a new biological virus that was destructive, we could detect it and sequence it in a day, create an antidote very quickly. That doesn't mean we can cross that off our worry list, because as soon as you put that into effect, it's obsolete just like software virus protection.
KURZWEILI take comfort from the fact that artificial intelligence is so well-integrated in the world. It doesn't just exist in a few dark government intelligence laboratories. It's on the belt of kids walking around in Africa. A kid in Africa with a Smartphone has access to more intelligent knowledge than the president of the United States did 15 years ago. There are over a billion smartphones and six billion cell phones, but technology is very widely distributed.
KURZWEILAnd I see that as a model of where we're headed. But are there dangers? Absolutely. Technology is a double-edged sword. I would discuss this a lot. I would make the point that we've benefited more than we have been harmed. Life expectancy was 37 in 1800. People were very poor. But we've also created a very destructive technology that could, in fact, potentially end all human or even all mammalian life.
REHMAll right, to Gainesville, FL. Good morning, Keith.
KEITHSuch a delight, this show, and thank you for having me. I'm struck by -- ask you for a prediction. Isaac Asimov spoke a lot about the, you know, the robotic intelligence and consciousness and he gave us some great stories about how to program within the systems some sense of rules that I guess we could call artificial intelligence morals. Do you see a need for that? And here in Gainesville, we have a lot of young upstart musician, engineer, technology companies and programmers and I have the privilege of working with them occasionally and talking about these things.
KEITHWhat are -- is there -- where is the programming to -- I don't want to say control, but to put some sense of judgment?
KURZWEILRight. Well, that's a very important issue. Ethics for robots is not just the providence of science fiction stories. We have intelligent machines that are making decisions that are life and death decisions, like, say, cruise missiles that decide where to go and make tactical decisions on their own using some form of intelligence and they're getting more and more independent and more and more intelligent.
KURZWEILYou know, who's -- right now, we manage to find some human biological human agent that's responsible if something goes wrong. But it's not going to be so clear cut if you go out to the 2030s. We're not going to be able to walk in a room and say, okay, humans on the left, computers on the right. It's going to be all mixed up. It's going to be hard to find a human who isn't, in some way, even if it's not inside their bodies and brains, enhanced by artificial intelligence.
KURZWEILAnd so who is responsible for what? You know, are there entities that are completely non-biological that we would hold responsible? They're different because they don't necessarily have one fixed body the way biological humans have. We have association of a physical person with the decisions that person makes. So you, for example, can put that physical person in jail, how do you do that with a software program?
KURZWEILBut that's where we're headed and it's a rich area of debate. And even before we have machines that are absolutely operating at human levels, it's important actually to discuss this issue.
REHMAre we, in some way then, as you suggests, headed for a kind of immortality?
KURZWEILWell, you know, in the 2030s, computers will have comparable levels of intelligence but not necessarily capturing your neo-cortex with all of its interconnections and patterns representing your personality skills and so on. But I do think actually that will become feasible, that's more like a 2040s or 2050 scenario. But we could actually back up the information in the brains. And when you back up your cell phone and you back up your notebook computer, you have information in your brain.
KURZWEILThat's not just metaphor. It's kind of frightening that it's not backed up. People do lose it. They lose all of it when they die. We can ultimately back that up and then recreate it. This immediately brings up philosophical issues if you, you know, bring back a back up. Is that the same person? Or is it just a new person with the same memories and personalities? But ultimately, we're going to become a hybrid of biological and non-biological thinking.
KURZWEILThe non-biological part will be backed up because everything we do that's non-biological is backed up. Ultimately, go out far enough, that will be where the action is. Most of our thinking will take place actually in the cloud and will be backed up. And that's a form of immortality, although I will point out that even information doesn't necessarily last forever. Try restoring some Word processing program -- a letter you wrote in WordStar 30 years ago and you'll see what I mean.
REHMAnd you're listening to "The Diane Rehm Show." You make me wonder about those who suffer now from Alzheimer's.
KURZWEILRight. Well, there's three reasons to reverse engineer the brain. One, we've been talking about, which is to create more intelligent machines based on the same methods, another is to understand ourselves. That's been the goal of the arts and sciences. Who are we? And an important part of that question is, you know, how do our brains work? But another reason is, in fact, just to do a better job fixing the brain.
KURZWEILNow we can already put computers in the human body and connect them into the brain, say, neural implants for Parkinson's disease. It's the latest generation. It allows you to download new software to this computer inside your body, connected into your brain from outside the patient. There was an article in the New York Times expressing concern about people hacking into that software. And the new ones are not just connected in one spot.
KURZWEILThey're now connected in hundreds and ultimately thousands of places. So they're getting more and more sophisticated. Today it requires surgery because it's not blood celled size, it's pea sized. But shrinking technologies and other exponential trend go out to the 2030s, we will routinely put these inside our bodies and brains. But our ability to fix neurological problems depends on our actually understanding how the brain works and how human dysfunction arises.
KURZWEILWhat's gone wrong in somebody with Alzheimer's or Parkinson's? And both through biotechnology and through understand neurology and neuroscience, we're making impressive gains in that. And I believe we'll actually be able to treat these diseases through a combination of these methods within the couple of decades ahead.
KURZWEILReverse engineer means to understand, to really understand how it works, which we have not been able to do up until just recently. Some of the best evidence I cited in the book came out in the weeks I was sending it off the publishing. In fact, we held it up -- I held up sending it to the publisher because some great new research came out. That's another exponential trend, the precision with which we can see inside the human brain.
KURZWEILWe can now see the individual connections being made in the living brain. We can see your brain create your thoughts. We can see your thoughts create your brain because of that feedback loop, that plasticity is part of the secret of human thinking.
REHMBut aren't you -- I mean, I know I am sort of thinking unconsciously most of the time?
KURZWEILWell, that's right. Most of what goes on in the neo-cortex and that is where we do our thinking is not conscious. We're only consciously aware of a small part of it and that actually get at the issue of responsibility and free will. You know, it's probably not a good defense to say, yes, I killed the victim but I wasn't paying attention and I wasn't conscious at what I was doing. So, you know, we are held responsible for what our brain does, whether we are consciously aware or not.
KURZWEILBut I cite some interesting studies where people actually begin to carry out a decision before they're even consciously aware that there was a decision to be made. And this was very, very interesting research. It's well-designed and people became aware that there was a decision after they already carried out the decision. So this brings up the mystery of consciousness. It's not just determining who and what is conscious, but even within our own brains what is it that we're conscious of?
KURZWEILThink of a walk you might have taken recently. How much of it do you remember? How much of it were you consciously aware of? You saw thousands of people and thousands of trees, but were you -- how much of it were you consciously aware of?
REHMI think the key to how to create a mind may be to be in the moment.
KURZWEILGreat advice to end the discussion on.
REHMRay Kurzweil, thank you so much.
REHMHis new book is titled, "How To Create A Mind: The Secret of Human Thought Revealed." Thanks for listening all. I'm Diane Rehm.
Most Recent Shows
Concerns over user privacy, the continued spread of misinformation and strong-arm tactics to crowd out competitors have users -- and governments -- rethinking their relationships with Facebook.
Why Diane's guest Ben Wittes says no.
In this moment of political discontent, when we talk of deep divides and a growing sense that our democracy has gone off track, historians counsel us to look to our…