The Singularity and Human Destiny by Patrick Tucker, Assistant Editor, THE FUTURIST In 1949, several decades before the birth of the PC, computer scientist John von Neumann an- nounced, “It would appear we have reached the limits of what is possible to achieve with com- puter technology.” He quickly and prudently added, “One should be careful with such state- ments; they tend to sound pretty silly in five years.” More than half a century later, it is Neumann’s caveat rather than his prediction that is borne out daily. Our computer intelligence is growing at an ever-quickening pace, surpassed only by our de- pendence on computer technology. For many of us, computers and computer-based devices have become not so much tools as appendages, third arms that are integral to our lives, cumbersome on occasion, and increasingly willful. This begs the obvious question: If computer power advances be- yond our control, how will that change us? In his most recent book, The Singularity Is Near, author and inventor Ray Kurzweil provides us with a clue. Imagine that with the help of a small device you could create a perfect replica of any ob- ject—Gianlorenzo Bernini’s David, the hub cap from a ’78 Dodge Dart—seemingly from vapor. Imagine that through virtual-reality software you won’t be shackled to a particular position in time, and could exist in several locations at once—work, home, a seaside bungalow in Bora Bora—and each setting in which you chose to locate yourself looked, sounded, and felt perfectly real. Imagine that you could live indefinitely in a world in which all poverty, pollution, and scarcity has been van- quished. Imagine that there existed no limitation to what you could do or be, except for those limita- tions that you imposed yourself. This scenario may sound like a cross between Thomas Moore’s Utopia and Christopher Marlow’s Doctor Faustus, but according to Ray Kurzweil it is our real and fast-approaching future. It will come about as the result of an ex- plosion in our technological abilities. We will in- corporate more computer-based processes into our biological functioning until we transcend our crude, earthly bodies entirely and become machine-based, virtually immortal. This coming period of rapid technological progress and its miraculous effects will occur within the next 50 years and is what Kurzweil refers to as the Singularity. The concept is both startling and optimistic, but it immediately provokes certain philosophi- cal concerns. If nanotechnology allows us to create any object, will any object ever again be valuable? What role will responsibility, temper- ance, and discipline play in a world where any urge can be gratified at almost the same moment it is felt? What will pass for morality when there is no mortal consequence to any action? These questions cannot and should not be an- swered all at once—either by Ray Kurzweil, his devotees, or his critics. Rather, what is im- portant is that they be asked, repeatedly and earnestly, and by as many people as possible. To further the debate of these key issues, THE FUTURIST presents Kurzweil’s insights and ideas along with invited commentaries from nanotechnology expert J. Storrs Hall, accelera- tion studies scholar John Smart, and soci- ologists Damien Broderick and Richard Eckersely. Together, they examine this issue of the Singularity to determine how near it is exactly, and explore what it might mean for humanity. Order the March-April 2006 issue of THE FUTURIST from https://www.wfs.org/futuristorder.htm.
Transcript
The Singularity and Human Destinyby Patrick Tucker, Assistant Editor, THE FUTURIST
In 1949, several decades before the birth of thePC, computer scientist John von Neumann an-nounced, “It would appear we have reached thelimits of what is possible to achieve with com-puter technology.” He quickly and prudentlyadded, “One should be careful with such state-ments; they tend to sound pretty silly in fiveyears.”
More than half a century later, it is Neumann’scaveat rather than his prediction that is borne outdaily. Our computer intelligence is growing at anever-quickening pace, surpassed only by our de-pendence on computer technology. For many ofus, computers and computer-based devices havebecome not so much tools as appendages, thirdarms that are integral to our lives, cumbersome onoccasion, and increasingly willful. This begs theobvious question: If computer power advances be-yond our control, how will that change us?
In his most recent book, The Singularity Is Near,author and inventor Ray Kurzweil provides uswith a clue. Imagine that with the help of a smalldevice you could create a perfect replica of any ob-ject—Gianlorenzo Bernini’s David, the hub capfrom a ’78 Dodge Dart—seemingly from vapor.Imagine that through virtual-reality software youwon’t be shackled to a particular position in time,and could exist in several locations at once—work,home, a seaside bungalow in Bora Bora—and eachsetting in which you chose to locate yourselflooked, sounded, and felt perfectly real. Imaginethat you could live indefinitely in a world in whichall poverty, pollution, and scarcity has been van-quished. Imagine that there existed no limitation towhat you could do or be, except for those limita-tions that you imposed yourself.
This scenario may sound like a cross between
Thomas Moore’s Utopia and ChristopherMarlow’s Doctor Faustus, but according to RayKurzweil it is our real and fast-approachingfuture. It will come about as the result of an ex-plosion in our technological abilities. We will in-corporate more computer-based processes intoour biological functioning until we transcendour crude, earthly bodies entirely and becomemachine-based, virtually immortal. This comingperiod of rapid technological progress and itsmiraculous effects will occur within the next 50years and is what Kurzweil refers to as theSingularity.
The concept is both startling and optimistic,but it immediately provokes certain philosophi-cal concerns. If nanotechnology allows us tocreate any object, will any object ever again bevaluable? What role will responsibility, temper-ance, and discipline play in a world where anyurge can be gratified at almost the same momentit is felt? What will pass for morality when thereis no mortal consequence to any action?
These questions cannot and should not be an-swered all at once—either by Ray Kurzweil,his devotees, or his critics. Rather, what is im-portant is that they be asked, repeatedly andearnestly, and by as many people as possible.
To further the debate of these key issues, THEFUTURIST presents Kurzweil’s insights andideas along with invited commentaries fromnanotechnology expert J. Storrs Hall, accelera-tion studies scholar John Smart, and soci-ologists Damien Broderick and RichardEckersely. Together, they examine this issue ofthe Singularity to determine how near it isexactly, and explore what it might mean forhumanity.
Order the March-April 2006 issue of THE FUTURIST from https://www.wfs.org/futuristorder.htm.
Author and inventor Ray Kurzweil sees aradical evolution of the human species in thenext 40 years.
We stand on the threshold of the most profoundand transformative event in the history of humanity,the “Singularity.”
What is the Singularity? From my perspective, theSingularity is a future period during which the paceof technological change will be so fast and far-reach-ing that human existence on this planet will beirreversibly altered. We will combine our brainpower—the knowledge, skills, and personality quirksthat make us human—with our computer power inorder to think, reason, communicate, and create inways we can scarcely even contemplate today.
This merger of man and machine, coupled with thesudden explosion in machine intelligence and rapidinnovation in gene research and nanotechnology, willresult in a world where there is no distinction betweenthe biological and the mechanical, or between physi-cal and virtual reality. These technological revolutionswill allow us to transcend our frail bodies with alltheir limitations. Illness, as we know it, will be eradi-cated. Through the use of nanotechnology, we will beable to manufacture almost any physical productupon demand, world hunger and poverty will besolved, and pollution will vanish. Human existencewill undergo a quantum leap in evolution. We will be
Reinventing Humanity:The Future of Machine–HumanIntelligence By Ray Kurzweil
able to live as long as we choose. Thecoming into being of such a world is,in essence, the Singularity.
How is it possible that we could beso close to this enormous changeand not see it? The answer is thequickening nature of technologicalinnovation. In thinking about thefuture, few people take into consid-eration the fact that human scientificprogress is exponential: It expandsby repeatedly multiplying by a con-stant (10 times 10 times 10, and soon) rather than linear (10 plus 10plus 10, and so on). I emphasize theexponential-versus-linear perspec-tive because it’s the most importantfailure that prognosticators make inconsidering future trends.
Our forebears expected what layahead of them to resemble what theyhad already experienced, with fewexceptions. Because they lived dur-ing a time when the rate of techno-logical innovation was so slow as tobe unnoticeable, their expectations ofan unchanged future were continu-ally fulfilled. Today, we have wit-nessed the acceleration of the curve.Therefore, we anticipate continuoustechnological progress and the socialrepercussions that follow. We see thefuture as being different from thepresent. But the future will be farmore surprising than most peoplerealize, because few observers havetruly internalized the implications ofthe fact that the rate of change isitself accelerating.
Exponential growth starts outslowly and virtually unnoticeably,but beyond the knee of the curve itturns explosive and profoundlytransformative. My models showthat we are doubling the paradigm-shift rate for technology innovationevery decade. In other words, thetwentieth century was graduallyspeeding up to today’s rate ofprogress; its achievements, therefore,were equivalent to about 20 years ofprogress at the rate of 2000. We’llmake another “20 years” of progressin just 14 years (by 2014), and thendo the same again in only sevenyears. To express this another way,we won’t experience 100 years oftechnological advance in the twenty-first century; we will witness on theorder of 20,000 years of progress(again, when measured by today’s
progress rate), or progress on a levelof about 1,000 times greater thanwhat was achieved in the twentiethcentury.
How Will We Know the Singularity Is Upon Us?
The first half of the twenty-firstcentury will be characterized bythree overlapping revolutions—ingenetics, nanotechnology, and robot-ics. These will usher in the beginningof this period of tremendous changeI refer to as the Singularity. We are inthe early stages of the genetics revo-lution today. By understanding theinformation processes underlyinglife, we are learning to reprogramour biology to achieve the virtualelimination of disease, dramatic ex-pansion of human potential, andradical life extension. However,Hans Moravec of Carnegie MellonUniversity’s Robotics Institute pointsout that, no matter how successfullywe fine-tune our DNA-based biol-ogy, biology will never be able tomatch what we will be able to engi-neer once we fully understand life’sprinciples of operation. In otherwords, we will always be “second-class robots.”
The nanotechnology revolutionwill enable us to redesign and re-build—molecule by molecule—ourbodies and brains and the worldwith which we interact, going far be-yond the limitations of biology.
But the most powerful impendingrevolution is the robotic revolution.By robotic, I am not referring ex-clusively—or even primarily—tohumanoid-looking droids that takeup physical space, but rather to arti-ficial intelligence in all its variations.
Following, I have laid out the prin-cipal components underlying each ofthese coming technological revolu-tions. While each new wave ofprogress will solve the problemsfrom earlier transformations, eachwill also introduce new perils. Each,operating both separately and inconcert, underpins the Singularity.
The Genetic Revolution
Genetic and molecular science willextend biology and correct its obvi-ous flaws (such as our vulnerabilityto disease). By the year 2020, the fulleffects of the genetic revolution willbe felt across society. We are rapidlygaining the knowledge and the toolsto drastically extend the usability ofthe “house” each of us calls his bodyand brain.
Nanomedicine researcher RobertFreitas estimates that eliminating50% of medically preventable condi-tions would extend human life ex-pectancy to 150 years. If we wereable to prevent 99% of naturally oc-curring medical problems, we’d liveto be more than 1,000 years old.
We can see the beginnings of thisawesome medical revolution today.The field of genetic biotechnology isfueled by a growing arsenal of tools.Drug discovery was once a matter offinding substrates (chemicals) thatproduced some beneficial resultwithout excessive side effects, aresearch method similar to earlyhumans’ seeking out rocks and othernatural implements that could beused for helpful purposes. Today, weare discovering the precise biochemi-cal pathways that underlie both dis-ease and aging processes. We are
PHOTOS.COM
In the future, more drugs will bedesigned to carry out precisemissions at the molecular level.
I first met Ray Kurzweil in 1999at a Foresight Institute meeting inPalo Alto. I was there to get somebackground information on nano-technology for a new book I waswriting. As I stood in the lunchline, a healthy-appearing man in
front of me was engaged in animated conversa-tion with a not nearly so healthy-looking secondman. Their topic of conversation was vitaminsand nutritional supplementation, a topic of greatinterest to me, a nutritionally oriented M.D.
I joined the conversation, and the healthy-looking man introduced himself as RayKurzweil. Ray and I continued our dialogue viae-mail after the conference ended, and a fewmonths later, he flew from his home in Boston toFrontier Medical Institute, my longevity clinic inDenver, for a comprehensive longevity medicalevaluation. We performed a comprehensive bat-tery of tests designed to uncover any health riskshe might still have so that together we could op-timize his already very sophisticated programfor health and longevity.
From the beginning, it was obvious that Raywould be a unique patient. I have many engi-neers as patients in my practice (and Ray is anengineer by training), so I am not surprisedwhen patients come to see me with a notebookof spreadsheets detailing various data ex-tracted from their daily lives: blood pressure,weight, cholesterol, blood sugar levels, amountof exercise, etc., carefully tabulated for severalyears. But all previous data collections I hadseen, even those organized into Excel andmeticulously graphed, paled in comparison toRay’s. His data collection was so thorough andmeticulous that he could tell me what he atefor lunch on June 23, 1989 (as well as what heate for lunch every other day for several yearsbefore that time). And not only what he ate,but the number of grams of each serving andcalories consumed, as well as the number ofcalories he burned that day through exercise—every day for decades!
As a result, it came as less of a surprise for me
to learn that Ray was taking over 200 supple-ment pills a day. Ray’s approach had been to ac-curately assess his personal health risks and thenquite simply to “reprogram his biochemistry.”Ongoing testing indicates that he is doing a re-markable job, as measurement of his biologicalage in my clinic indicates that he is now almosttwo decades younger than his chronological age,and all of his health risks appear under optimalcontrol.
Ray was already working on his new book, TheSingularity Is Near, at that time, and I had justcompleted my first book, The Baby Boomers’ Guideto Living Forever. It was natural that our e-mail di-alogue moved into discussion of the prospectsfor truly radical life extension for people of allages, including older boomers like ourselves. Asour e-mails multiplied into the many thousands,we decided to organize the information and seeif we had the makings of a new book that wewould co-author. I created a preliminary table ofcontents, Ray organized the information fromour e-mails, and, another 10,000 e-mails or solater, our joint book, Fantastic Voyage: Live LongEnough to Live Forever, was written in the midst ofRay’s writing of The Singularity Is Near.
Ray felt that he was writing these books to-gether as a unit and that there was synergybetween them. The Singularity Is Near detailsRay’s vision of the astounding possibilities of theworld of the near future as the Singularity un-folds sometime within the next few decades. InFantastic Voyage, we provide readers with the in-formation they need to live long enough andremain healthy enough to fully experience thewonders of life in the post-Singularity world. Inwriting these two books, Ray has painted a clearpicture of the future and provided a blueprint forhow to get there.
About the AuthorTerry Grossman, M.D., is the founder and medical directorof Frontier Medical Institute in Denver, Colorado. He is the co-author of Fantastic Voyage: The Science BehindRadical Life Extension with Ray Kurzweil. Telephone 303-233-4247; Web site www.fmiclinic.com.
Ray Kurzweil’s Plan for Cheating DeathA cure for aging may be found in the next fifty years. The trick now is to l ive
long enough to be there when it happens. Here’s how Kurzweil is trying to do it.
able to design drugs to carry out pre-cise missions at the molecular level.With recently developed gene tech-nologies, we’re on the verge of beingable to control how genes expressthemselves. Gene expression is theprocess by which cellular compo-nents (specifically RNA and the ribo-somes) produce proteins accordingto a precise genetic blueprint. Whileevery human cell contains a com-plete DNA sample, and thus the fullcomplement of the body’s genes, aspecific cell, such as a skin cell or apancreatic islet cell, gets its charac-teristics from only the fraction ofgenetic information relevant to thatparticular cell type.
Gene expression is controlled bypeptides (molecules made up of se-quences of up to 100 amino acids)and short RNA strands. We are nowbeginning to learn how theseprocesses work. Many new therapiescurrently in development and testingare based on manipulating peptideseither to turn off the expression ofdisease-causing genes or to turn ondesirable genes that may otherwisenot be expressed in a particular typeof cell. A new technique called RNAinterference is able to destroy themessenger RNA expressing a geneand thereby effectively turn thatgene off.
Accelerating progress in biotech-nology will enable us to reprogramour genes and metabolic processes topropel the fields of genomics (influ-encing genes), proteomics (under-standing and influencing the role ofproteins), gene therapy (suppressinggene expression as well as addingnew genetic information), rationaldrug design (formulating drugs thattarget precise changes in disease andaging processes), as well as the ther-apeutic cloning of rejuvenated cells,tissues, and organs.
The Nanotechnology Revolution
Nanotechnology promises thetools to rebuild the physical world,our bodies, and our brains, molecu-lar fragment by molecular fragmentand potentially atom by atom. Weare shrinking the key features (work-ing parts), in accordance with thelaw of accelerating returns, at an ex-
ponential rate (over four per lineardimension per decade, or about 100per 3-D volume.) At this rate, thekey feature sizes for most electronicand many mechanical technologieswill be in the nanotechnologyrange—generally considered to beless than 100 nanometers (one-billionth of one meter)—by the2020s. Electronics has alreadydipped below this threshold, al-though not yet in three-dimensionalstructures and not yet in structuresthat are capable of assembling othersimilar structures, an essential stepbefore nanotechnology can reach itspromised potential. Meanwhile,rapid progress has been made re-cently in preparing the conceptualframework and design ideas for thecoming age of nanotechnology.
Nanotechnology has expanded toinclude any technology in which amachine’s key features are measuredby fewer than 100 nanometers. Justas contemporary electronics has al-ready quietly slipped into this nanorealm, the area of biological andmedical applications has already en-tered the era of nanoparticles, inwhich nanoscale objects are beingdeveloped to create more-effectivetests and treatments.
In the area of testing and diagno-sis, nanoparticles are being em-ployed in experimental biologicaltests as tags and labels to greatly en-hance sensitivity in detecting sub-stances such as proteins. Magneticnanotags can be used to bind withantibodies that can then be read us-ing magnetic probes while still in-side the body. Successful experi-ments have been conducted withgold nanoparticles that are bound to
DNA segments and can rapidly testfor specific DNA sequences in asample. Small nanoscale beadscalled quantum dots can be pro-grammed with specific codes com-bining multiple colors, similar to acolor bar code, that can facilitatetracking of substances through thebody.
In the future, nanoscale deviceswill run hundreds of tests simultane-ously on tiny samples of a given sub-stance. These devices will allow ex-tensive tests to be conducted onnearly invisible samples of blood.
In the area of treatment, a particu-larly exciting application of this tech-nology is the harnessing of nanopar-ticles to deliver medication tospecific sites in the body. Nanopar-ticles can guide drugs into cell wallsand through the blood-brain barrier.Nanoscale packages can be designedto hold drugs, protect them throughthe gastrointestinal tract, ferry themto specific locations, and then releasethem in sophisticated ways that canbe influenced and controlled, wire-lessly, from outside the body.
Nanotherapeutics in Alachua,Florida, has developed a biodegrad-able polymer only several nanome-ters thick that uses this approach.Meanwhile, scientists at McGill Uni-versity in Montreal have demon-strated a nanopill with structures inthe 25 to 45 nanometer range. Thenanopill is small enough to passthrough the cell wall and delivermedications directly to targetedstructures inside the cell.
MicroCHIPS of Bedford, Massa-chusetts, has developed a computer-ized device that is implanted underthe skin and delivers precise mix-tures of medicines from hundreds ofnanoscale wells inside the device.Future versions of the device are ex-pected to be able to measure bloodlevels of substances such as glucose.The system could be used as an arti-ficial pancreas, releasing preciseamounts of insulin based on theblood glucose response. The systemwould also be capable of simulatingany other hormone-producing or-gan. If trials go smoothly, the systemcould be on the market by 2008. An-other innovative proposal is to guidenanoparticles (probably composed ofgold) to a tumor site and then heat
42 THE FUTURIST March-April 2006 www.wfs.org
J. STORRS HALL
A nanoscale, self-replicating robot orutility foglet could join together with otherfoglets to form a solid wall that wouldchange in shape and appearance as de-sired by the user. Foglet technology willallow for full-immersion virtual-reality envi-ronments by the 2030s.
them with infrared beams to destroythe cancer cells.
The revolution in nanotechnologywill allow us to do a great deal morethan simply treat disease. Ultimately,nanotech will enable us to redesignand rebuild not only our bodies andbrains, but also the world withwhich we interact. The full realiza-tion of nanotechnology, however,will lag behind the biotechnologyrevolution by about one decade. Butby the mid to late 2020s, the effectsof the nanotech revolution will bewidespread and obvious.
Nanotechnology and The Human Brain
The most important and radicalapplication particularly of circa-2030nanobots will be to expand ourminds through the merger of bio-logical and nonbiological, or “ma-chine,” intelligence. In the next 25years, we will learn how to augmentour 100 trillion very slow inter-neuronal connections with high-speed virtual connections via nano-robotics. This will allow us to greatlyboost our pattern-recognition abili-
ties, memories, and overallthinking capacity, as well asto directly interface withpowerful forms of computerintelligence. The technologywill also provide wirelesscommunication from onebrain to another.
In other words, the age oftelepathic communication isalmost upon us.
Our brains today are rela-tively fixed in design. Al-though we do add patternsof interneuronal connec-tions and neurotransmitterconcentrations as a normalpart of the learning process,the current overall capacity of thehuman brain is highly constrained.As humanity’s artificial-intelligence(AI) capabilities begin to upstage ourhuman intelligence at the end of the2030s, we will be able to move be-yond the basic architecture of thebrain’s neural regions.
Brain implants based on massivelydistributed intelligent nanobots willgreatly expand our memories andotherwise vastly improve all of oursensory, pattern-recognition, andcognitive abilities. Since the nanobotswill be communicating with one an-other, they will be able to create anyset of new neural connections, breakexisting connections (by suppressingneural firing), create new hybrid bio-logical and computer networks, andadd completely mechanical net-works, as well as interface intimatelywith new computer programs andartificial intelligences.
The implementation of artificial in-telligence in our biological systemswill mark an evolutionary leap for-ward for humanity, but it also im-plies we will indeed become more“machine” than “human.” Billions ofnanobots will travel through thebloodstream in our bodies andbrains. In our bodies, they will de-stroy pathogens, correct DNA errors,eliminate toxins, and perform manyother tasks to enhance our physicalwell-being. As a result, we will beable to live indefinitely withoutaging.
In our brains, nanobots will inter-act with our biological neurons. Thiswill provide full-immersion virtualreality incorporating all of the
senses, as well as neurological corre-lates of our emotions, from withinthe nervous system. More impor-tantly, this intimate connectionbetween our biological thinking andthe machine intelligence we are cre-ating will profoundly expandhuman intelligence.
Warfare will move toward nanobot-b a s e d w e a p o n s , a s w e l l a scyberweapons. Learning will firstmove online, but once our brains arefully online we will be able to down-load new knowledge and skills. Therole of work will be to createknowledge of all kinds, from musicand art to math and science. The roleof play will also be to createknowledge. In the future, therewon’t be a clear distinction betweenwork and play.
The Robotic Revolution
Of the three technological revolu-tions underlying the Singularity(genetic, nano-mechanical, and ro-botic), the most profound is roboticor, as it is commonly called, thestrong artificial intelligence revolution.This refers to the creation of com-puter thinking ability that exceedsthe thinking ability of humans. Weare very close to the day when fullybiological humans (as we now knowthem) cease to be the dominant intel-ligence on the planet. By the end ofthis century, computational or me-chanical intelligence will be trillionsof trillions of times more powerfulthan unaided human brain power. Iargue that computer, or as I call itnonbiological intelligence, should still
THE FUTURIST March-April 2006 www.wfs.org 43
PHOTOS.COM
ILLUSTRATION: PATRICK TUCKER, WFS / IMAGES: PHOTOS.COM
Our brains today are relatively fixed in de-sign. As humanity’s artificial-intelligence (AI)capabilities begin to upstage our human in-telligence at the end of the 2030s, we willbe able to move beyond the basic architec-ture of the brain’s neural regions. But artifi-cial Intelligence will be based, at least inpart, on a human-made version of a fullyfunctional human brain.
be considered human since it is fullyderived from human–machine civi-lization and will be based, at least inpart, on a human-made version of afully functional human brain. Themerger of these two worlds of intelli-gence is not merely a merger of bio-logical and mechanical thinkingmediums, but also (and more impor-tantly) a merger of method and orga-nizational thinking that will expandour minds in virtually every imagi-nable way.
Biological human thinking is lim-ited to 1016 calculations per second(cps) per human brain (based onneuromorphic modeling of brainregions) and about 1026 cps for allhuman brains. These figures will notappreciably change, even with bio-engineering adjustments to ourgenome. The processing capacity ofnonbiological intelligence or strongAI, in contrast, is growing at an ex-ponential rate (with the rate itselfincreasing) and will vastly exceedbiological intelligence by the mid-2040s.
Artificial intelligence will neces-sarily exceed human intelligence forseveral reasons. First, machines canshare knowledge and communicatewith one another far more efficientlythan can humans. As humans, we donot have the means to exchange thevast patterns of interneuronal con-nections and neurotransmitter-con-centration levels that comprise ourlearning, knowledge, and skills,other than through slow, language-based communication.
Second, humanity’s intellectualskills have developed in ways thathave been evolutionarily encouragedin natural environments. Thoseskills, which are primarily based onour abilities to recognize and extractmeaning from patterns, enable us tobe highly proficient in certain tasks,such as distinguishing faces, identi-fying objects, and recognizing lan-guage sounds. Unfortunately, ourbrains are less well-suited for deal-ing with more-complex patterns,such as those that exist in financial,scientific, or product data. The appli-cation of computer-based techniqueswill allow us to fully master pattern-recognition paradigms. Finally, ashuman knowledge migrates to theWeb, machines will demonstrate
increased proficiency in reading, un-derstanding, and synthesizing allhuman–machine information.
The Chicken or the Egg
A key question regarding the Sin-gularity is whether the “chicken”(strong AI) or the “egg” (nanotech-nology) will come first. In otherwords, will strong AI lead to fullnanotechnology (molecular-manu-facturing assemblers that can turninformation into physical products),or will full nanotechnology lead tostrong AI?
The logic of the first premise isthat strong AI would be in a positionto solve any remaining design prob-lems required to implement fullnanotechnology. The second premiseis based on the assumption thathardware requirements for strong AIwill be met by nanotechnology-based computation. Likewise, thesoftware requirements for engineer-ing strong AI would be facilitated bynanobots. These microscopic ma-chines will allow us to create highlydetailed scans of human brainsalong with diagrams of how thehuman brain is able to do all thewonderful things that have longmystified us, such as create meaning,contextualize information, and expe-rience emotion. Once we fully un-derstand how the brain functions,we will be able to recreate the phe-nomenon of human thinking in ma-chines. We will endow computers,already superior to us in the per-formance of mechanical tasks, withlifelike intelligence.
Progress in both areas (nano androbotic) will necessarily use ourmost-advanced tools, so advances ineach field will simultaneously facili-tate the other. However, I do expectthat the most important nanotechno-logical breakthroughs will emergeprior to strong AI, but only by a fewyears (around 2025 for nanotech-nology and 2029 for strong AI).
As revolutionary as nanotech-nology will be, strong AI will havefar more profound consequences.Nanotechnology is powerful but notnecessarily intelligent. We can deviseways of at least trying to manage theenormous powers of nanotech-nology, but superintelligence by its
nature cannot be controlled.The nano/robotic revolution will
also force us to reconsider the verydefinition of human. Not only willwe be surrounded by machines thatwill display distinctly human char-acteristics, but also we will be lesshuman from a literal standpoint.
Despite the wonderful future po-tential of medicine, real humanlongevity will only be attained whenwe move away from our biologicalbodies entirely. As we move towarda software-based existence, we willgain the means of “backing our-selves up” (storing the key patternsunderlying our knowledge, skills,and personality in a digital setting)thereby enabling a virtual immortal-ity. Thanks to nanotechnology, wewill have bodies that we can notonly modify, but also change intonew forms at will. We will be able toquickly change our bodies in full-im-mersion virtual-reality environmentsincorporating all of the senses dur-ing the 2020s and in real reality inthe 2040s.
Implications of the Singularity
What will be the nature of humanexperience once computer intelli-gence predominates? What are theimplications for the human–machinecivilization when strong AI andnanotechnology can create any prod-uct, any situation, any environmentthat we can imagine at will? I stressthe role of imagination here becausewe will still be constrained in ourcreations to what we can imagine.But our tools for bringing imagina-tion to life are growing exponentiallymore powerful.
People often go through threestages in considering the impact offuture technology: awe and wonder-ment at its potential to overcomeage-old problems, then a sense ofdread at the grave new dangers thataccompany these novel technologies,followed finally by the realizationthat the only viable and responsiblepath is to set a careful course thatcan realize the benefits while manag-ing the dangers.
My own expectation is that thecreative and constructive applica-tions of these technologies will dom-inate, as I believe they do today.
However, we need to vastly increaseour investment in developing spe-cific defensive technologies. We areat the critical stage where we need todirectly implement defensive tech-nologies for nanotechnology duringthe late teen years of this century.
I believe that a narrow relinquish-ment of the development of certaincapabilities needs to be part of ourethical response to the dangers oftwenty-first-century technologicalchallenges. For example, Bill Joy andI wrote a joint op-ed piece in the NewYork Times recently, criticizing thepublication of the 1918 flu genomeon the Web because it constitutes adangerous blueprint. Another con-structive example of this are theethical guidelines proposed by theForesight Institute: namely, that nano-technologists agree to relinquish the
development of physical entities thatcan self-replicate in a natural envi-ronment free of any human controlor override mechanism. However,deciding in favor of too many limita-tions and restrictions would under-mine economic progress and isethically unjustified, given the op-portunity to alleviate disease, over-come poverty, and clean up theenvironment.
We don’t have to look past todayto see the intertwined promise andperil of technological advancement.Imagine describing the dangers(atomic and hydrogen bombs, forone thing) that exist today to peoplewho lived a couple of hundred yearsago. They would think it mad to takesuch risks. But how many people in2006 would really want to go back tothe short, brutish, disease-filled,
poverty-stricken, disaster-prone livesthat 99% of the human race strug-gled through two centuries ago?
We may romanticize the past, butup until fairly recently most of hu-manity lived extremely fragile livesin which one all-too-common mis-fortune could spell disaster. Twohundred years ago, life expectancyfor females in the record-holdingcountry (Sweden) was roughly 35years—very brief compared with thelongest life expectancy today, almost85 years for Japanese women. Lifeexpectancy for males was roughly 33years, compared with the current 79years. Half a day was often requiredto prepare an evening meal, andhard labor characterized mosthuman activity. There were no socialsafety nets. Substantial portions ofour species still live in this precari-
ILLUSTRATION: PATRICK TUCKER, WFS / IMAGES: PHOTOS.COM
ous way, which is at least one reasonto continue technological progressand the economic improvement thataccompanies it. Only technology,with its ability to provide orders ofmagnitude of advances in capabilityand affordability, has the scale toconfront problems such as poverty,disease, pollution, and the otheroverriding concerns of society today.The benefits of applying ourselves tothese challenges cannot be over-stated.
As the Singularity approaches, wewill have to reconsider our ideasabout the nature of human life andredesign our human institutions. In-telligence on and around Earth willcontinue to expand exponentiallyuntil we reach the limits of matterand energy to support intelligentcomputation. As we approach thislimit in our corner of the galaxy, theintelligence of our civilization willexpand outward into the rest of the
universe, quickly reaching the fastestspeed possible. We understand thatspeed to be the speed of light, butthere are suggestions that we may beable to circumvent this apparentlimit (conceivably by taking short-cuts through “wormholes,” or hypo-thetical shortcuts through space andtime).
A common view is that science hasconsistently been correcting ouroverly inflated view of our own sig-nificance. Stephen Jay Gould said,“The most important scientific revo-lutions all include, as their only com-mon feature, the dethronement ofhuman arrogance from one pedestalafter another of previous convictionsabout our centrality in the cosmos.”
Instead, it turns out we are central.Our ability to create models—virtualrealities—in our brains, combinedwith our modest-looking thumbs,has been sufficient to usher in an-other form of evolution: technology.
That development enabled the per-sistence of the accelerating pace thatstarted with biological evolution. Itmay continue until the entire uni-verse is at our fingertips.
About the AuthorRay Kurzweil is a scientist,inventor, and entrepreneur.He has received 12 hon-orary doctorates in science,engineering, music, and hu-mane letters from Rensse-laer Polytechnic Institute,
Hofstra University, and other leading col-leges and universities. He has been in-ducted into the National Inventors Hall ofFame and received the 1999 NationalMedal of Technology, among numerousother awards. His Web site, www.KurzweilAI.net, has more than a million readers andincludes a free daily e-newsletter. Thisarticle draws, in part, on his most recentbook, The Singularity Is Near: WhenHumans Transcend Biology (Viking, 2005).
I have a few differences of opinion with Kurzweilabout the coming Singularity.
I think he is being overly optimistic about biotechnol-ogy’s ability to create substantially better biologicalhuman beings. While we’ll certainly learn to pushhuman capacities to their natural limits in comingdecades, I see nothing on the horizon that would allowus to exceed those limits. Biology seems far too frail,slow, complex, and well defended (both at the molecularlevel and with regard to social custom) for that to beplausible within any reasonable time frame. Further-more, by the time we are able to substantially improveour biology, we probably won’t want to, as there will befar more interesting and powerful technological environ-ments available to us instead. This points to the impor-tance of understanding the relative accelerations of vari-ous technologies (in this case, biological vs.technological).
In his book, Kurzweil makes a major contribution tothe literature on acceleration studies by clearly explain-ing technological acceleration curves. These accelerationcurves show that the longer we use a technology, themore we get out of it: We use less energy, space, andtime, and we get more capacity for less cost. Technologi-cal acceleration curves are a little-understood area, butthanks to pioneers like Kurzweil, interest and research inthe field are advancing.
The common notion that the “futurecan’t be predicted” is demonstrably falsewith regard to a wide number of accelerat-ing physical-computational trends, eventhough we do not yet know specifically how those tech-nologies will be implemented. We can no longer ignorethe profound technological changes occurring all aroundus.
It’s also time we acknowledged the slowness of humanbiology compared with our technological progeny. Ourmachines are increasingly exceeding us in the perform-ance of more and more tasks, from guiding objects likemissiles or satellites to assembling other machines. Theyare merging with us ever more intimately and are learn-ing how to reconfigure our biology in new and signifi-cantly faster technological domains.
Something very interesting is happening, and humanbeings are selective catalysts, not absolute controllers, ofthis process. Let us face this openly, and investigate it ac-tively, so that we may guide these developments aswisely as possible.
About the AuthorJohn Smart is a developmental systems theorist and the presidentof the Acceleration Studies Foundation, 2227 Amirante, San Pedro,California 90732. Telephone 310-831-4191; Web site www.accelerating.org.
Some years ago, I reviewed Ray Kurzweil’s earlier book,The Age of Spiritual Machines, for the Foresight Nanotech In-stitute’s newsletter. Shortly thereafter I met him in personat a Foresight event, and he had read the review. He toldme, “Of all the people who reviewed my book, you werethe only one who said I was too conservative!”
The Singularity Is Near is very well researched, and ingeneral, Kurzweil’s predictions are about as good as it’spossible to get for things that far in advance. I still thinkhe’s too conservative in one specific area: Synthetic com-puter-based artificial intelligence will become availablewell before nanotechnology makes neuron-level brainscans possible in the 2020s.
What’s happening is that existing technologies likefunctional MRI are beginning to give us a high-levelfunctional block diagram of the brain’s processes. At thesame time, the hardware capable of running a strong, ar-tificially intelligent computer, by most estimates, is herenow, though it’s still pricey.
Existing AI software techniques can build programsthat are experts in any well-defined field. The break-throughs necessary for such programs to learn for them-selves could happen easily in the next decade—one ortwo decades before Kurzweil predicts.
Kurzweil finesses the issue of runaway AI by propos-ing a pathway where machine intelligence is patterned
after human brains, so that they wouldhave our morals and values built in. In-deed, this would clearly be the wise andprudent course. Unfortunately, it seems alltoo likely that a shortcut exists without that kind of safe-guard. Corporations already use huge computer systemsfor data mining and decision support that employ so-phisticated algorithms no human manager understands.It’s a very short step to having such a system make betterdecisions than the managers do, as far as the corpora-tion’s bottom line is concerned.
The Singularity may mean different things to differentpeople. To me, it is that point where intelligences signifi-cantly greater than our own control so many of the es-sential processes that figure in our lives that merehumans can’t predict what happens next. This futuremay be even nearer than Ray Kurzweil has predicted.
About the AuthorJ. Storrs Hall is chief scientist of Nanorex Inc. and a fellow of theMolecular Engineering Research Institute. His address is NanorexInc., 2738 Turtle Ridge Drive, Bloomfield Hills, Michigan 48302.
He is the author of Nanofuture: What’s Next for Nanotechnology(Prometheus Books, 2005, $28), which is available from the FuturistBookshelf, www.wfs.org/bkshelf.htm.
Runaway Artificial Intelligence? By J. Storrs Hall
A quarter century ago, we’d have laughed at theprospect of “Dick Tracy” cell phones with cameras; nowthey’re everywhere, and nobody noticed after the firstfew days. So the jump to the idea of a Singularity is notreally extraordinary. But should we really expect evermore substantial changes to follow the same accelerat-ing, headlong pace?
It’s reasonable to expect affordable computers to be smallerand more powerful, 1,000 times improved in a decade, onemillion times in 20 years, one billion in 30. By then, some ma-chines might have capabilities to rival the human mind. Anew intelligent species might share the planet with us.
In addition, developing technologies such as molecu-lar manufacture—nanotechnology—will allow the veryengines of productivity to be copied cheaply and distrib-uted widely. If that happens the gap between rich andpoor might diminish. However, it will only occur if wefind ways to prevent portable nanofactories from makinglethal weapons available to any child or psychopath.We’ll be able to solve most of the problems that currentlyvex us—global warming (to the extent that it’s caused byhumans), water and food shortages, provision of clean,cheap power, and so on.
There is a scary downside that I discussed nearly adecade ago in my book The Spike: Dirt-cheap molecularmanufacturing may end poverty and strife, but there ex-ists a risk that a world of lotus-eaters will degenerate
into gang wars among those for whom liferetains no discipline or meaning outside ofarbitrary local status and violence. People(young men especially) with full belliesgained effortlessly, but lacking meaning in their lives,often find purpose in ganging up on each other in fits ofmurderous primate chest-pounding. Making Huxleiansoma, or “feelies,” the opiate of the people might help,but that, too, is a sickening prospect.
On the other hand, those strictly unforeseeable andmysterious changes captured in the word “Singularity”are likely to overwhelm and surpass such predictabledownsides of any technological utopia or dystopia. Theeeriest aspect of accelerating change is that we ourselves,and our children, will be the ones soaking in it. Thesooner we start thinking seriously about the prospect, thebetter prepared we’ll be.
About the AuthorDamien Broderick is a senior fellow in the Department of Englishand Cultural Studies at the University of Melbourne, Australia. Hisfuturist books include The Spike (1997, rev. 2001), The Last MortalGeneration (1999), and Ferocious Minds (2005). His novels dealingwith the Singularity include Transcension (2002), The Hunger ofTime (with Rory Barnes, 2003), Godplayers (2005), and K-Ma-chines (forthcoming, 2006). He lives in Melbourne, Australia, andSan Antonio, Texas.
Nanofactories, Gang Wars, and “Feelies” By Damien BroderickBETH GWINN
I have sometimes asked audiences if they are inspiredor excited by the sort of techno-utopian vision repre-sented by the Singularity; almost no one is. In my sur-veys over the past decade, I found dwindling minoritiesof young people (one-fifth to one-quarter) believe in thesort of technical fixes to human problems that RayKurzweil champions, while an increased majority (aboutthree-quarters) believe science and technology are alien-ating people from each other and from nature.
The question I ask is, why? Why pursue this future? Idon’t pose this question dismissively, or derogatorily, butout of genuine curiosity and a desire for an open, honestconversation. I’m skeptical of arguments that say pre-technological humans led short, nasty, and brutish lives.Yes, life expectancy was lower—mainly because of high
rates of infant mortality—but those whosurvived often lived socially and spiritu-ally rich lives. It doesn’t make evolution-ary sense to believe humans lived in mis-ery until we discovered technological progress. Animalsin the wild don’t live that way, and humans have been,for most of their history, animals in the wild.
The future world that Kurzweil describes bears almostno relationship to human well-being that I am aware of.In essence, human health and happiness comes frombeing connected and engaged, from being suspended ina web of relationships and interests—personal, social,and spiritual—that give meaning to our lives. The inti-macy and support provided by close personal relation-ships seem to matter most; isolation exacts the highest
price. The need to belong is more im-portant than the need to be rich.Meaning matters more than moneyand what it buys.
We are left with the matter of des-tiny: It is our preordained fate,Kurzweil suggests, to advance tech-nologically “until the entire universeis at our fingertips.” The questionthen becomes, preordained bywhom or what? Biological evolutionhas not set this course for us. Is tech-nology itself the planner? Perhaps itwill eventually be, but not yet. IsGod the entity doing the ordaining?A lot of religious people would havesomething to say about that, and arelikely to strenuously, and even vio-lently, oppose what the Singularitypromises, as I have argued before(THE FUTURIST, November-December2001).
We are left to conclude that wewill do this because it is we whohave decided it is our destiny. Butwe have made no such decision,really, as the observations withwhich I began this commentaryshow.
About the AuthorRichard Eckersley is a fellow at the NationalCentre for Epidemiology and PopulationHealth at the Australian National University,Canberra, Australia, and author of Well &Good: Morality, Meaning and Happiness(Text Publishing, Melbourne, 2005). [email protected].
FEEDBACK: Send your commentsabout this article to [email protected].
Techno-Utopia and Human Values By Richard Eckersley
Ray Kurzweil reponds:
Richard Eckersley’s idyllic notion ofhuman life hundreds of years ago beliesour scientific knowledge of history. Twohundred years ago, there was no under-standing of sanitation, so bacterial in-fections were rampant. There were noantibiotics and no social safety nets, soan infectious disease was a disasterplunging a family into desperation.Thomas Hobbes’s characterization in1651 of human life as solitary, poor,nasty, brutish, and short was on themark. Even ignoring infant mortality,life expectancy was in the 30s only acouple of hundred years ago. Schubert’s and Mozart’s deaths at31 and 35 respectively were typical.
Eckersley bases his romanticized idea of ancient life on com-munication and the relationships fostered by communication.But much of modern technology is directed at just this basichuman need. The telephone allowed people to be together evenif far apart geographically. The Internet is the quintessentialcommunication technology. Social networks and the panoply ofnew ways to make connection are creating communities basedon genuine common interests rather than the accident ofgeography.
This decentralized electronic communication is also highly de-mocratizing. In a book I wrote in the mid-1980s, I predicted thedemise of the Soviet Union from the impact of the then-emergingcommunication networks, and that is indeed what happened inthe early 1990s. The democracy movement we saw in the 1990sand since is similarly fueled by our unprecedented abilities tostay in touch.
If Eckersley really sticks to his own philosophy, he won’t bearound for very long to influence the debate. But I hope that hewill take advantage of the life extension—and enhancement—technologies that will emerge in the decades ahead, so that wecan continue this dialogue through this century and beyond.
