 |
|
|
|
|
|
| |
Origin >
The Singularity >
Accelerating Intelligence: Where Will Technology Lead Us?
Permanent link to this article: http://www.kurzweilai.net/meme/frame.html?main=/articles/art0450.html
Printable Version |
|
| |
|
Accelerating Intelligence: Where Will Technology Lead Us?
Kurzweil gave a Special Address at BusinessWeek's The Digital Economy New Priorities: Building A Collaborative Enterprise In Uncertain Times conference on December 6, 2001 in San Francisco. He introduced business CEOs to the Singularity -- the moment when distinctions between human and machine intelligence disappear.
Published March 26, 2002
A video of the talk is now online at:
http://www.kurzweilai.net/articles/videos/ray_bizweek.ram
Excerpts from Ray Kurzweil's remarks:
"In considering the genesis of Moore's Law, I put 49 famous
computing devices over the past century on an exponential graph.
From this exercise, it became apparent that the acceleration of
computing power did not start with integrated circuits, but has
continued through multiple paradigm shifts(electromechanical calculators,
relays, vacuum tubes, transistors, and finally integrated circuits).
"Moore's Law was not the first, but the fifth paradigm, to
provide exponential growth in computing. The next paradigm, which
will involve computing in three dimensions rather than the two manifested
in today's flat chips, will lead to computing at the molecular,
and ultimately the subatomic level. We can be confident that the
acceleration of computing will survive the well-anticipated demise
of Moore' s Law.
"There are comparable exponential trends underlying a wide
variety of other technologies: communications (both wired and wireless),
brain scanning speeds and resolutions, genome scanning, and miniaturization
(we are currently shrinking technology at a rate of 5.6 per linear
dimension per decade). Even the rate of technological progress is
speeding up, now doubling each decade. The mathematical models I've
developed over the past couple of decades to describe these trends,
which I call the law of accelerating returns, has proven predictive
of the developments we've seen during the 1990s. From these models,
I believe we can be confident of continued exponential growth in
these and other technologies for the foreseeable future.
"By 2009, computers will disappear. Displays will be written
directly onto our retinas by devices in our eyeglasses and contact
lenses. In addition to virtual high-resolution displays, these intimate
displays will provide full immersion visual virtual reality. We
will have ubiquitous, very-high-bandwidth wireless connection to
the Internet at all times. "Going to a web site" will
mean entering a virtual reality environment -- at least for the
visual and auditory sense -- where we will meet other real people.
There will be simulated people as well, but the virtual personalities
will not be up to human standards, at least not by 2009. The electronics
for all of this will be so small that it will be invisibly embedded
in our glasses and clothing.
"By 2029, as a result of continuing trends in miniaturization,
computation, and communication, we will have billions of nanobots
-- intelligent robots the same of blood cells or smaller -- traveling
through the capillaries of our brain communicating directly with
our biological neurons. By taking up positions next to every nerve
fiber coming from all of our senses, the nanobots will provide full-immersion
virtual reality involving all five of the senses. So we will enter
virtual reality environments (via the web, of course) of our choice
and meet people, both real and virtual, only now the difference
won't be so clear.
"Just as people today beam their images from little web cams
out onto the Internet for others to share, many people in 2029 will
beam the full stream of signals coming directly from their senses
onto the web. We will then be able to experience what other people
are experiencing, à la John Malkovich. Of course, the everyday
lives of many such experience beamers may not be all that compelling,
so there will be plenty of prerecorded experiences we can plug into
it. Beyond just the five senses, these shared experiences will include
emotional responses, sexual pleasure, and other mental reactions.
"Brain implants based on these distributed intelligent nanobots
will extend our brains in every conceivable way, massively expanding
our memory and otherwise vastly improving all of our sensory, pattern-recognition
and cognitive abilities.
"Oh, and one more thing: we'll live a long time too. The expanding
human life span is another one of those exponential trends. In the
eighteenth century, we added a few days every year to human longevity;
during the nineteenth century we added a couple of weeks each year;
and now we're adding almost a half a year every year. With the revolutions
in rational drug design, genomics, therapeutic cloning of our own
organs and tissues, and related developments in bio-information
sciences, we will be adding more than a year every year within ten
years. So take care of yourself the old-fashioned way for just a
little while longer, and you may actually get to experience the
remarkable century ahead."
| | Join the discussion about this article on Mind·X! | |
|
| |
Mind·X Discussion About This Article:
 |
|
|
 |
Re: My goal
|
|
 |
|
My goal in being a programmer is to help develop something that will put me out of business =] By this, I of course mean AI. I'm hoping all *need* for business and wealth aquisition will also go the way of the Dodo.
Jeremy> Of course, when this happens, who knows what else will go the way of the dodo.
Anything else will be purely for the sake of broadening one's horizons. A world of unlimited, free intellectual and 'spiritual' (artistic, whatever) pursuits is my definition of Utopia.
Jeremy>Some folks may continue their intellectual pursuits, but why? Anything humans can do, AIs will be doing better... making your "broadening horizons" look rather narrow.
When people like you put yourself out of a job (the rest of us too) we will soon be outpaced by thinking machines. Sure, we will have all our needs met, but but any control over how that is done will be lost forever. Lazy, greedy, smart but lacking wisdom, and trapped in our golden cage. |
|
 |
|
 |
|
|
|
|
Re: Accelerating Intelligence: Where Will Technology Lead Us?
|
|
|
|
In my view, the ultimate purpose of any human being is the pursue of knowledge and understanding about the world, by study, exploration, sharing and creation of models and frameworks, applications, paintings, art, papers, literature, etc. This is what we should do all the time 'this is the ultimate source of happiness.
Sometimes I wonder if we really need an economy like the one we have to pursue these goals. I guess we need a little bit more of cultural evolution.
|
|
|
|
|
|
|
|
|
Re: Accelerating Intelligence: Where Will Technology Lead Us?
|
|
|
|
If we place ourselves as observers, outside the evolutionary process, and look at the destiny of matter, self-organizing agents, life, and ultimately the universe, I would say that an agent in a system will be successful if maximizes its prospects according to the medium. Nevertheless, let's remember that the agent is part of the system. Considering this, the total outcome would be an increased level of efficiency and order (entropy). A clear conclusion then is that life and its evolved prospects are the force that will have a glorious impact on the faith of the universe. It is certainly impossible to predict the nature of this 'destiny', but I like the idea of increasing level of awareness and conciseness that would arise as this organic/technological/pattern explosive, ever expanding, intelligence mass creates as it develops.
Concerning the different levels of awareness and evolutionary states that co-exist, I think it is perfectly possible for one level to be completely ignorant of the other 'think bacteria and a human being. In a less dramatic example we can imagine a pet -- let's say a dog -- and its master; they cooperate and synergize, nevertheless they operate in different levels of awareness. Taking this exercise a level up, it is perfectly reasonable then to acknowledge that a vast number of levels exist above us. Why we do not see them? Maybe we do, and we are not conscious of it -- probably it is not even ethical for them to proceed and be recognized in some way or another.
In the short term, technology will allow us to bite our tails ' to modify, enhance and expand our 'program' -- by the combined usage of biology, nanotechnology and cybernetics. This creates yet another self-referential loop, which outcome will carry us to a new level of awareness. No matter if we become mecha or enhanced orga or both, we will communicate better, live longer, and create new levels of order. At the individual level one thing remains: The satisfaction ' think happiness -- in seeking, exploring, creating and expressing.
Pablo Troncoso - PabTro |
|
|
|
|
|
|
|
|
Re: Accelerating Intelligence: Where Will Technology Lead Us?
|
|
|
|
Dear Pablo,
Certainly, you can ascribe certain general characteristics to successful lifeforms, but it is less easy to explain why they become extinct. By all measures, the dinosaurs were a very successful and vibrant group of creatures, but one day a very unlucky hit by a big, old rock put an end to them. Similarly, technology has, in general, been the best of friends to the human race, providing us with great amounts of food and comfort and breeding opportunities. But there is that nagging problem of 50,000 or so nuclear warheads ready and waiting to do their thing to turn the earth into a blackened wasteland. And there is the scary and not too distant prospect that an unenlightened and only moderately talented genetic engineer might do a little mixing and matching of viral genomes to produce a cross between, say, a cold virus and an ebola virus to create a malevalent monster that mother nature would never dream of. This one bad guy might be able to undo the good work of tens of thousands of good and industrious genetic engineers in their efforts to help mankind. It's just like the single bad day the dinosaurs had 65 million years ago. The millions of previous sunny and warm days in the Cretaceous told them that they had it made in the shade. But it didn't work out that way. It seems to me that we are throwing a lot of very big rocks around in the neighborhood of the human race and one is bound to come crashing back: nuclear warfare, grey ooze or viral bombs - any one and many others can give rise to that very bad day that the dinosaurs experienced.
I hope we make it. But the lesson of the dinosaurs tells us that our survival isn't guaranteed. And speaking of faith, I for one hope and pray that the Creator made this huge, long-lasting universe to propagate love. After all, physicists tell us it is easy to change a few universal constants and create a universe that self destructs promptly, with no hope for life. Why waste everyone's time with a universe which produces sentient beings only if they are doomed to blow themselves up? We've gotta have faith and hope, but now more than ever is the time to study ethics and learn the basis for doing the right thing for our fellow sentient beings. Pray for peace!
Tim Dean |
|
|
|
|
|
|
|
|
Re: Accelerating Intelligence: Where Will Technology Lead Us?
|
|
|
|
I agree with Tim Dean, the near future holds some very perilous threats! Biowarfare is a real possibility (inevitablity?) that can really slow things down!
However, if we make it past these threats, the accelerating intelligence thing is sort of troublesome as well. I doubt that once we ascended to god status, we would stop growing, the cyle time to second generation gods might be VERY short indeed. So far, each successive leap of evolution has been shorter and shorter. The Apes lasted longer than our immediate ancestors, and the amphibians that were way before us lasted a very long time.
2 possibilites i see:
The reason we have not been contacted by other races, is that just before they get smart enough to contact us, they get smart enough to destroy themselves with terrible weapons.
Or if the doubling time gets rapidly shorter each evolutionary leap, once we ascend to the next level, it may be weeks until they next hundred levls of evolution happen, and the 100th power gods are so powerful, they upset the cosmic balance and........BOOM, the big bang. Game over- start again! |
|
|
|
|
|
|
|
|
Re: Accelerating Intelligence: Where Will Technology Lead Us?
|
|
|
|
The English language is wonderful. It is very elastic. However, its elasticity is also sometimes a problem. English vocabulary is sometimes somewhat vague.
Take the word "god", for instance. The word is used to define many different concepts...from mythological creatures akin to fairies and sprites to a philosophical shortand for ultimate reality.
In my own work, I try to avoid using the word "god" for that very reason. If I have to describe the religioius views of a group of people, I import the appropriate foreign word, and try my damndest to define it, so that my prose is as clear as possible.
We see the same problem here. When we talk of possible future AI entities, and we describe them as "godlike", what do we mean? Do we mean the cool, calm God of Plato and Aristotle? Do we mean the angry Yahweh of the Old Testament? Or the lustful and all-too-human denizens of Mount Olympus? Do we mean the remote creator god of the Yoruba?
I propose the banning of the word "god" from these discussions, unless the word is used in a religious context.
BC
|
|
|
|
|
|
|
|
|
Re: Accelerating Intelligence: Where Will Technology Lead Us?
|
|
|
|
Commie, (mind if I call you Marx or Lenin? :-))
I remember when I was a kid (way, way back when), there was an article in the newspaper that predicted we would all own a helicopter and use it to go to work by the end of the 20th Century.
It is easy to ridicule the predictions of the 20th Century--but if we ridicule where they were too optimistic, fairness demands that we look at where they were too pessimistic, too.
In many respects, life in the US in 2002 is more like 1952 than 1952 was like 1902. In 1902, heavier-than-air flight had not taken place, radio was only an experiment, automobiles were hobbyist's toys, and the major causes of death (in the US) were various infectious diseases.
In 1952, jet airplanes were in the sky, television was omnipresent, nearly every family in the US had a car, and the major causes of death in the US were heart disease and cancer.
Futurists in the period between the 1950s and 1970s extrapolated the developments of the last few decades forward. Men landed on the moon in 1969. It was only natural for those futurists to predict that Mars would be just around the corner.
But the futurists didn't look at the economic reality. We don't have colonies on the Moon because there is, at least for now, no good economic reason to have them. There has been no manned Mars mission because nobody has put forth a decent reason for spending the money.
Meanwhile, futurists of the 60s and 70s didn't anticipate the omnipresence of computer technology that we see today. In 1976, to build a computer with the memory of the one that I'm using right now would have cost nearly a half a million dollars. I bought this G4 Macintosh last year for less than $2000. And in 1976, even with the memory, the processing speed wouldn't have been available. In the '50s and '60s, this would have seemed more incredible than colonies on the moon.
Today, futurists continue to extrapolate current realities forward. I am a middle-aged man (hell, honestly in most of human history, I would be considered OLD), and I can safely predict that most of the futurists' predictions will be wrong. But on the other hand, I am sure that there will be developments that nobody has forseen.
BC
|
|
|
|
|
|
|
|
|
Re: Accelerating Intelligence: Where Will Technology Lead Us?
|
|
|
|
>Commie, (mind if I call you Marx or Lenin? :-))
Of course not :)
>Futurists in the period between the 1950s and 1970s extrapolated the developments of the last few decades forward.
Now don't get me wrong, I'm not trying to poke holes (as Colby stated before), but the people making predictions today are in a similar situation to those making them in the 1970s. The foundation of the futurists predictions also lies in the events of the past few decades. Take for examples, the fact that computers have gone from huge, expensive, and slow to small, cheap, and excessively fast and the fact that we have just cracked the human genome; obviously if computers keep advancing at the current rate, then the near future holds computers that are infinitely accelerating with a conscience of their own (AI). Likewise, a whole new life awaits human beings: an all-curing panacea, inexpensive genetic manipulations, and body-repair kits that keep humans living for centuries, right? WRONG. I'm not trying to say that these things will never come to be, I'm simply saying that they will not appear as soon as futurists are expecting them. The predictions of the 1970s faced problems that either hindered their emergence or even kept them from coming into existance, problems such as economic difficulties, lack of reason to make the prediction come to life, or simply the insufficient amount of human intelligence carry out the prediction.
Now tell me this, do we have the slightest idea of how to physically create a "brain", or some sort of AI unit? We're far from it, so how can you expect Singularity within a few decades? Until someone figures out a way to create a hovercar, I wouldn't even bother arguing that we are to face Singularity in the near future. Yes, the future holds very many interesting surprises and new inventions that probably you or I could have never imagined, but some of the futurists ideas are way too erratic.
-Commie (Marx, Lenin, or whatever you want to call me :P ) |
|
|
|
|
|
|
|
|
Re: Accelerating Intelligence: Where Will Technology Lead Us?
|
|
|
|
Kurzweil makes some interesting points about the trends in technology and how the different sectors are growing in these exponential curve patterns. However, he seems to make them out to be a great predictor of what is going to happen, and bases his claims on the trends in these curves. Although this may be true for certain ‘big item’ electronics, as is plotted on the curve, predicting what these big items are going to be is a whole other problem. By ‘big item’ I mean something that the public will mass adopt, such as the television, the Internet, the telephone and so on.
Kurzweil says that by 2009, computers will disappear completely and displays will be written directly onto our retinas by devices in our eyeglasses and contact lenses. He expects this to provide us with full immersion visual virtual reality. Although this sounds like something that would be very promising, I don’t believe it is practical for the public at all. First, many people are now getting laser eye surgery to remove those annoying contact lenses and glasses from their faces. Personally, I don’t like the idea of having something sitting on my face while I am browsing the Internet. And there’s also the problem of interacting with our glasses. Do we have to carry around a portable keyboard that connects to our glasses somehow? That does not seem like a much better idea than a laptop. Also, the idea of being visually immersed in my email or banking site is not very appealing. Taking off my glasses when someone taps me on the shoulder to show me something is not my idea of convenience. It’s no wonder nearly all virtual reality devices have failed, including television glasses.
By 2029, Kurzweil mentions that miniaturization will be so far ahead that we will have nanobots inside our bodies traveling through our bloodstreams and living in the capillaries of our brain. This would be a great product if it ever came to realization. However, I’m not so sure the public would react kindly to having themselves injected with little machines; especially little machines that can connect to the Internet. There have been so many viruses and bugs associated with software and the Internet that the public adoption of such a thing would be minimal at best. The public just doesn’t have enough trust in software or machinery to trust their lives to them. I do agree though, that in a purely medical sense, these bots would be excellent as long as they did not connect to the public Internet. However, if the public did happen to adopt these nanobots, it is quite scary to imagine what may happen. On the downside, people may be ‘hacked’ into and taken control of or simply killed. On the upside, we can potentially become more knowledgeable than ever before. But if these nanobots can make us feel whatever we want to feel, then why become smarter? Instead, I think people will just tell the nanobots to make them feel happy all the time. What could be better than that? We will likely all just become zombies doing nothing but being happy until we eventually die. Unless of course these nanobots are effected by magnets and there are some mischievous animals around.
Personally, I think technology is going to lead us to the ultimate question of whether or not we have a soul/spirit/life force/etc. When we have finally reverse engineered the brain and decoded our DNA, we will be able to build a robot to replicate their effects on our body completely. If at that point, the robot is able to learn without us telling it anything, enjoy music, feel emotions and interact in our society, then I believe we have answered the question. If we ever reach that point, then perhaps all we are is exactly like the robot, a sequence of predictable reactions like a giant physics experiment. In that case, everything we have ever done is all fated to be and we don’t have any instincts and we can’t really think. This is not an outcome I would like see as I am on the side of Descartes; “I think, therefore I am”. But in either case, I do believe technology will answer the question for us.
|
|
|
|
|
|
|
|
