In this Technology Review article, Raymond Kurzweil and Michael Dertouzos debate Bill Joy’s Wired article urging “relinquishment” of research in certain risky areas of nanotechnology, genetics, and robotics.
Originally published January 1, 2001 in High Technology. Published on KurzweilAI.net March 7, 2001.
Although I agree with Michael Dertouzos’ conclusion in rejecting Bill Joy’s prescription to relinquish “our pursuit of certain kinds of knowledge,” I come to this view through a very different route. Although I am often paired with Bill Joy as the technology optimist versus Bill’s pessimism, I do share his concerns about the dangers of self-replicating technologies. Michael is being shortsighted in his skepticism.
Michael writes that “just because chips…are getting faster doesn’t mean they’ll get smarter, let alone lead to self-replication.” First of all, machines are already “getting smarter.” As just one of many contemporary examples, I’ve recently held conversations with a person who speaks only German by translating my English speech in real time into human-sounding German speech (by combining speech recognition, language translation and speech synthesis) and similarly converting their spoken German replies into English speech. Although not perfect, this capability was not feasible at all just a few years ago. The intelligence of our technology does not need to be at human levels to be dangerous. Second, the implication that self-replication is harder than intelligence is not accurate. Software viruses, although not very intelligent, are self-replicating as well as being potentially destructive. Bioengineered biological viruses are not far behind. As for nanotechnology-based self-replication, that’s further out, but the consensus in that community is this will be feasible in the 2020s, if not sooner.
Many long-range forecasts of technical feasibility in future time periods dramatically underestimate the power of future technology because they are based on what I call the “intuitive linear” view of technological progress rather than the “historical exponential” view. When people think of a future period, they intuitively assume that the current rate of progress will continue for the period being considered. However, careful consideration of the pace of technology shows that the rate of progress is not constant, but it is human nature to adapt to the changing pace, so the intuitive view is that the pace will continue at the current rate. It is typical, therefore, that even sophisticated commentators, when considering the future, extrapolate the current pace of change over the next 10 years or 100 years to determine their expectations. This is why I call this way of looking at the future the “intuitive linear” view.
But any serious consideration of the history of technology shows that technological change is at least exponential, not linear. There are a great many examples of this, including exponential trends in computation, communication, brain scanning, miniaturization and multiple aspects of biotechnology. One can examine this data in many different ways, on many different time scales and for a wide variety of different phenomena, and we find (at least) double exponential growth, a phenomenon I call the “law of accelerating returns.” The law of accelerating returns does not rely on an assumption of the continuation of Moore’s law, but is based on a rich model of diverse technological processes. What it clearly shows is that technology, particularly the pace of technological change, advances (at least) exponentially, not linearly, and has been doing so since the advent of technology. That is why people tend to overestimate what can be achieved in the short term (because we tend to leave out necessary details) but underestimate what can be achieved in the long term (because exponential growth is ignored).
This observation also applies to paradigm shift rates, which are currently doubling (approximately) every decade. So the technological progress in the 21st century will be equivalent to what would require (in the linear view) on the order of 20,000 years.
Michael’s argument that we cannot always anticipate the effects of a particular technology is irrelevant here. These exponential trends in computation and communication technologies are greatly empowering the individual. Of course, that’s good news in many ways. These trends are behind the pervasive trend we see toward democratization, and are reshaping power relations at all levels of society. But these technologies are also empowering and amplifying our destructive impulses. It’s not necessary to anticipate all of the ultimate uses of a technology to see that there is danger in, for example, every college biotechnology lab having the ability to create self-replicating biological pathogens.
However, I do reject Joy’s call for relinquishment of broad areas of technology (such as nanotechnology) despite my not sharing Michael’s skepticism on the feasibility of these technologies. Technology has always been a double-edged sword. We don’t need to look any further than today’s technology to see this. If we imagine describing the dangers that exist today (enough nuclear explosive power to destroy all mammalian life, just for starters) to people who lived a couple of hundred years ago, they would think it mad to take such risks. On the other hand, how many people in the year 2001 would really want to go back to the short, brutish, disease-filled, poverty-stricken, disaster-prone lives that 99 percent of the human race struggled through a couple of centuries ago?
People often go through three stages in examining the impact of future technology: awe and wonderment at its potential to overcome age-old problems, then a sense of dread at a new set of grave dangers that accompany these new technologies, followed, finally and hopefully, by the realization that the only viable and responsible path is to set a careful course that can realize the promise while managing the peril.
The continued opportunity to alleviate human distress is one important motivation for continuing technological advancement. Also compelling are the already apparent economic gains, which will continue to hasten in the decades ahead. There is an insistent economic imperative to continue technological progress: relinquishing technological advancement would be economic suicide for individuals, companies and nations.
Which brings us to the issue of relinquishment, which is Bill Joy’s most controversial recommendation and personal commitment. Forgoing fields such as nanotechnology is untenable. Nanotechnology is simply the inevitable end result of a persistent trend toward miniaturization that pervades all of technology. It is far from a single centralized effort but is being pursued by a myriad of projects with many diverse goals.
Furthermore, abandonment of broad areas of technology will only push them underground, where development would continue unimpeded by ethics and regulation. In such a situation, it would be the less stable, less responsible practitioners (for example, the terrorists) who would have all the expertise.
The constructive response to these dangers is not a simple one: It combines professional ethical guidelines (which already exist in biotechnology and are currently being drafted by nanotechnologists), oversight by regulatory bodies and the development of technology-specific “immune” responses, as well as computer-assisted surveillance by law enforcement organizations. As we go forward, balancing our cherished rights of privacy with our need to be protected from the malicious use of powerful 21st-century technologies will be one of many profound challenges.
Technology will remain a double-edged sword, and the story of the 21st century has not yet been written. It represents vast power to be used for all humankind’s purposes. We have no choice but to work hard to apply these quickening technologies to advance our human values, despite what often appears to be a lack of consensus on what those values should be.
In my column, I observed that we have been incapable of judging where technologies are headed, hence we should not relinquish a new technology, based strictly on reason. Ray agrees with my conclusion, but for a different reason: He sees technology growing exponentially, thereby offering us the opportunity to alleviate human distress and hasten future economic gains. From his perspective, my point is “irrelevant,” and my views on the future of technology are “skeptical.” Let’s punch through to the underlying issues, which are vital, for they point at a fundamental and all-too-often ignored relationship between technology and humanity.
Ray’s exponential-growth argument is half the story: No doubt, the number of transistors on a chip has grown and will continue to grow for a while. But transistors and the systems made with them are used by people. And that’s where exponential change stops! Has word-processing software, running on millions of transistors, empowered humans to contribute better writings than Socrates, Descartes or Lao Tzu?
Technologies have undergone dramatic change in the last few centuries. But people’s basic needs for food, shelter, nurturing, procreation and survival have not changed in thousands of years. Nor has the rapid growth of technology altered love, hate, spirituality or the building and destruction of human relationships. Granted, when we are in the frying pan, surrounded by the sizzling oil of rapidly changing technologies, we feel that everything around us is accelerating. But, from the longer range perspective of human history and evolution, change is far more gradual. The novelty of our modern tools is counterbalanced by the constancy of our ancient needs.
As a result, technological growth, regardless of its magnitude, does not automatically empower us. It does so only when it matches our ability to use it for human purposes. And that doesn’t happen as often as we’d like. Just think of the growing millions of AIDS cases in Africa, beyond our control. Or, in the industrial world, ask yourself whether we are truly better off surrounded by hordes of complex digital devices that force us to serve them rather than the other way around.
Our humanity meets technology in other ways, too: In forecasting the future of technology, Ray laments that most people use “linear thinking” that builds on existing patterns, thereby missing the big “nonlinear” ideas that are the true drivers of change. Once again, this is only half the story: In the last three decades, as I witnessed the new ideas and the 50-some startups that arose from the MIT Laboratory for Computer Science, I observed a pattern: Every successful technological innovation is the result of two simultaneous forces-a controlled insanity needed to break away from the stranglehold of current reason and ideas, and a disciplined assessment of potential human utility, to filter out the truly absurd. Focusing only on the wild part is not enough: Without a check, it often leads to exhibitionistic thinking, calculated to shock. Wild ideas can be great. But I draw a hard line when such ideas are paraded in front of a lay population as inevitable, or even likely.
That is the case with much of the futurology in today’s media, because of the high value we all place on entertainment. With all the talk about intelligent agents, most people think they can go buy them in the corner drugstore. Ray, too, brings up his experience with speech translation to demonstrate computer intelligence. The Lab for Computer Science is delightfully full of Victor Zue’s celebrated systems that can understand spoken English, Spanish and Mandarin, as long as the context is restricted, for example to let you ask about the weather, or to book an airline flight. Does that make them intelligent? No. Conventionally, “intelligence” is centered on our ability to reason, even imperfectly, using common sense. If we dub as intelligent, often for marketing or wishful-thinking purposes, every technological advance that mimics a tiny corner of human behavior, we will be distorting our language and exaggerating the virtues of our technology. We have no basis today to assert that machine intelligence will or will not be achieved. Stating that it will go one way or the other is to assert a belief, which is fine, as long as we say so. Does this mean that machine intelligence will never be achieved? Certainly not. Does it mean that it will be achieved? Certainly not. All it means is that we don’t know-an exciting proposition that motivates us to go find out.
Attention-seizing, outlandish ideas are easy and fun to concoct. Far more difficult is to pick future directions that are likely. My preferred way for doing this, which has served me well, though not flawlessly, for the last 30 years, is this: Put in a salad bowl the wildest, most forward-thinking technological ideas that you can imagine. (This is the craziness part.) Then add your best sense of what will be useful to people. (That’s the rational part.) Start mixing the salad. If you are lucky, something will pop up that begins to qualify on both counts. Grab it and run with it, since the best way to forecast the future is to build it. This forecasting approach combines “nonlinear” ideas with the “linear” notion of human utility, and with a hopeful dab of serendipity.
Ray observes that technology is a double-edged sword. I agree, but I prefer to think of it as an axe that can be used to build a house or chop the head off an adversary, depending on intentions. The good news is that since the angels and the devils are inside us, rather than within the axe, the ratio of good to evil uses of a technology is the same as the ratio of good to evil people who use that technology…which stays pretty constant through the ages. Technological progress will not automatically cause us to be engulfed by evil, as some people fear.
But for the same reason, potentially harmful uses of technology will always be near us, and we will need to deal with them. I agree with Ray’s suggestions that we do so via ethical guidelines, regulatory overviews, immune response and computer-assisted surveillance. These, however, are partial remedies, rooted in reason, which has repeatedly let us down in assessing future technological directions. We need to go further.
As human beings, we have a rational, logical dimension, but also a physical, an emotional and a spiritual one. We are not fully human unless we exercise all of these capabilities in concert, as we have done throughout the millennia. To rely entirely on reason is to ascribe omniscience to a few ounces of meat, tucked inside the skull bones of antlike creatures roaming a small corner of an infinite universe–hardly a rational proposition! To live in this increasingly complex, awesome and marvelous world that surrounds us, which we barely understand, we need to marshal everything we’ve got that makes us human.
This brings us back to the point of my column, which is also the main theme of this discussion: When we marvel at the exponential growth of an emerging technology, we must keep in mind the constancy of the human beings who will use it. When we forecast a likely future direction, we need to balance the excitement of imaginative “nonlinear” ideas with their potential human utility. And when we are trying to cope with the potential harm of a new technology, we should use all our human capabilities to form our judgment.
To render technology useful, we must blend it with humanity. This process will serve us best if, alongside our most promising technologies, we bring our full humanity, augmenting our rational powers with our feelings, our actions and our faith. We cannot do this by reason alone!