{\rtf1{\info{\title Technological Singularity}{\author Vernor Vinge}}\ansi\ansic pg1252\deff0\deflang1033 {\fonttbl{\f0\froman

\fprq2\fcharset128 Times New Roman;}{\f1\froman\fprq2\fchars et128 Times New Roman;}{\f2\fswiss\fprq2\fcharset128 Arial;}{\f3\fnil\fprq2\fcha rset128 Arial;}{\f4\fnil\fprq2\fcharset128 MS Mincho;}{\f5\fnil\fprq2\fcharset12 8 Tahoma;}{\f6\fnil\fprq0\fcharset128 Tahoma;}} {\stylesheet{\ql \li0\ri0\nowidctlpar\wrapdefault\faauto\rin0\lin0\itap0 \rtlch\ fcs1 \af25\afs24\alang1033 \ltrch\fcs0 \fs24\lang1033\langfe255\cgrid\langnp1033 \langfenp255 \snext0 Normal;} {\s1\ql \li0\ri0\sb240\sa120\keepn\nowidctlpar\wrapdefault\faauto\outlinelevel0\ rin0\lin0\itap0 \rtlch\fcs1 \ab\af0\afs32\alang1033 \ltrch\fcs0 \b\fs32\lang1033 \langfe255\loch\f1\hich\af1\dbch\af26\cgrid\langnp1033\langfenp255 \sbasedon15 \ snext16 \slink21 heading 1;} {\s2\ql \li0\ri0\sb240\sa120\keepn\nowidctlpar\wrapdefault\faauto\outlinelevel1\ rin0\lin0\itap0 \rtlch\fcs1 \ab\ai\af0\afs28\alang1033 \ltrch\fcs0 \b\i\fs28\lan g1033\langfe255\loch\f1\hich\af1\dbch\af26\cgrid\langnp1033\langfenp255 \sbasedo n15 \snext16 \slink22 heading 2;} {\s3\ql \li0\ri0\sb240\sa120\keepn\nowidctlpar\wrapdefault\faauto\outlinelevel2\ rin0\lin0\itap0 \rtlch\fcs1 \ab\af0\afs28\alang1033 \ltrch\fcs0 \b\fs28\lang1033 \langfe255\loch\f1\hich\af1\dbch\af26\cgrid\langnp1033\langfenp255 \sbasedon15 \ snext16 \slink23 heading 3;} {\s4\ql \li0\ri0\sb240\sa120\keepn\nowidctlpar\wrapdefault\faauto\outlinelevel3\ rin0\lin0\itap0 \rtlch\fcs1 \ab\ai\af0\afs23\alang1033 \ltrch\fcs0\b\i\fs23\lang 1033\langfe255\loch\f1\hich\af1\dbch\af26\cgrid\langnp1033\langfenp255 \sbasedon 15 \snext16 \slink24 heading 4;} {\s5\ql \li0\ri0\sb240\sa120\keepn\nowidctlpar\wrapdefault\faauto\outlinelevel4\ rin0\lin0\itap0 \rtlch\fcs1 \ab\af0\afs23\alang1033 \ltrch\fcs0 \b\fs23\lang1033 \langfe255\loch\f1\hich\af1\dbch\af26\cgrid\langnp1033\langfenp255 \sbasedon15 \ snext16 \slink25 heading 5;} {\s6\ql \li0\ri0\sb240\sa120\keepn\nowidctlpar\wrapdefault\faauto\outlinelevel5\ rin0\lin0\itap0 \rtlch\fcs1 \ab\af0\afs21\alang1033 \ltrch\fcs0 \b\fs21\lang1033 \langfe255\loch\f1\hich\af1\dbch\af26\cgrid\langnp1033\langfenp255 \sbasedon15 \ snext16 \slink26 heading 6;}} Magazine: Whole Earth Review{ Issue: December 10, 1993\par\pard\plain\hyphpar}{ Title: Technological Singularity\par\pard\plain\hyphpar}{ Author: Vernor Vinge\par\pard\plain\hyphpar}{ Technological Singularity\par\pard\plain\hyphpar}{ by Vernor Vinge\par\pard\plain\hyphpar}{ Vernor Vinge's vision of a technological "singularity" in humanity's near futur e has haunted me since I first read of it in his science-fiction novel, Marooned in Realtime (1986). I'm persuaded that the acceleration of technology-accelerat ion is even now distorting human institutions and expectations, whether or not w e are approaching a metaphorical "event horizon" beyond which everything becomes unrecognizable. When I invited Vinge to write something about his current views on the singularity for the recent issue of Whole Earth Review that I guest-edit ed, he replied that he had just presented a paper on the subject for the VISION21 Symposium, sponsored by the NASA Lewis Research Center and the Ohio Aerospace Institute. In due course he revised the piece and sent it along. I can think of no other technical paper that has so many references to science-fiction literat ure, as well it should. Vinge is a mathematician at San Diego State University, specializing in distributed computing and computer architecture. One of his shor t stories, "True Names" (1981), is often mentioned along with John Brunner's Sho ckwave Rider and William Gibson's Neuromancer as an inspiration to the current g eneration of online computer pioneers. Vinge's two "Realtime" novels (combined i n Across Realtime --1991) have been nominated for Hugo Awards, science fiction's top prize. His new novel, A Fire Upon the Deep, won the 1993 Hugo; it's reviewe d on p. 95. \par\pard\plain\hyphpar}{ --Stewart Brand\par\pard\plain\hyphpar}{ --------------------------------------------------------------------------TECHN

It is a point where our old models must be discarded and a new reality rules. In fact. Yet when it finally ha ppens. Progress in hardware has followed an amazingly steady curve in the last few decades. we can solve many problems thousa nds of times faster than natural selection could. The fi rst three possibilities depend on improvements in computer hardware. (Charles Platt has pointed out that AI en thusiasts have been making claims like this for thirty years. Developments th at were thought might only happen in "a million years" (if ever) will likely hap pen in the next century. never properly absorbed. We are on the edge of change comparable to the rise of human life on E arth. In the 1950s very few saw it: Stan Ulam 1 paraphrased John von Neumann as saying:\par\pard\plain\ hyphpar}{ One conversation centered on the ever-accelerating progress of technology and c hanges in the mode of human life. (To date. Since the design of mac . With out that we would get a glut of technical riches. by creating the means to execute those simulations at much higher speeds.) The 1960s saw recognition of some of the implications of superhuman intelligence. Computer/human interfaces may become so intimate that us ers may reasonably be considered superhumanly intelligent. --Vernor Vinge 1. including this notice. could not continue. J. We humans have the ability to internalize the world and conduct what-if's in our heads. Science may achieve this breakth rough by several means (and this is another reason for having confidence that th e event will occur): Computers that are "awake" and superhumanly intelligent may be developed. Now. let me be more specific: I'll be surprised if this event occurs before 2005 or after 2030. But if the answer is "yes." then there is l ittle doubt that more intelligent beings can be constructed shortly thereafter. Von Neumann even uses the term singularity.) Large computer networks and their associated users may "wake up" as superhumanl y intelligent entities. The best analogy I see is to the ev olutionary past: Animals can adapt to problems and make inventions. Good wrote:\par\pard\plain\hyphpar}{ Let an ultraintelligent machine be defined as a machine that can far surpass al l the intellectual activities of any man however clever. there has been much controversy as to whether we can cr eate human equivalence in a machine. \p ar\pard\plain\hyphpar}{ This change will be a throwing-away of all the human rules.)\par\pard\plain\hyphpar}{ What are the consequences of this event? When greater-than-human intelligence d rives progress. March 30-31. 1993. Based on this trend. as we know them.OLOGICAL SINGULARITY\par\pard\plain\hyphpar}{ (c) 1993 by Vernor Vinge (This article may be reproduced for noncommercial purp oses if it is copied in its entirety. I. I believe that the creation of greater-than-human intelligence wi ll occur during the next thirty years. \par\pard\plain\hyphp ar}{ Biological science may provide means to improve natural human intellect. the superhumanity is the essence of the Singularity. which gives the appearance of approaching some essential singularity in the history of the race beyond which human affairs. (For me. t hough it appears he is thinking of normal progress. It's fair to call this event a singularity ("the Singul arity" for the purposes of this piece). Just so I'm not gu ilty of a relative-time ambiguity. there seems no r eason why progress itself would not involve the creation of still more intellige nt entities --on a still-shorter time scale. we are entering a regime as rad ically different from our human past as we humans are from the lower animals. perhaps in the blin k of an eye --an exponential runaway beyond any hope of control. The precise cause of this change is the imminent creation by technology of entities with greater-than-human intelligence. not the creation of superhum an intellect. it may still be a great surprise and a greater unknown. a point that will loom vaster and vaster o ver human affairs until the notion becomes a commonplace. that progress will be much more rapid.) A slightly differe nt version of this article was presented at the VISION-21 Symposium sponsored by NASA Lewis Research Center and the Ohio Aerospace Institute. but often no faster than natural selection can do its work --the world acts as its own simul ator in the case of natural selection. What Is The Singularity? \par\pard\plain\hyphpar}{ The acceleration of technological progress has been the central feature of this century.

an ultraintelligent machine could design even better machines. as we slide toward the edge? How will the approach of the Singularity spread across the human world view? For a while yet. If networking is widespread enough (into ubiquitous embedded systems). the "hard" science-fiction writers are the ones who try to write specific stories about all that technology may do for us. . until we have hardware as powerful as a human brain it is probably foolish to think we'll be able to c reate human-equivalent (or greater) intelligence. If so. then the arrival of self-aware machines will not happen until after th e development of hardware that is substantially more powerful than humans' natur al equipment. there would then unquestionably be an "intelligenc e explosion. it will probably occur faster than any technical revolution seen so far. And fo r all my technological optimism. an ultrain telligent machine will be built and that it will be the last invention that man need make. involving lots of false starts and experimentatio n. we should see more symptoms."). Once. In August 1992. The precipitating event will likely be unexpected --perhaps even by the researchers involved ("But all our pr evious models were catatonic! We were just tweaking some parameters . it may seem as if our artifacts as a whole had suddenly awakened. (I have heard thoug htful comicbook writers worry about how to create spectacular effects when every thing visible can be produced by the technologically commonplace. . In the coming of the Singularity. Thus the fir st ultraintelligent machine is the last invention that man need ever make. Thinking Machines Corporation held a workshop to inves . We have tools right now (symbo lic math programs. Put an other way: the work that is truly productive is the domain of a steadily smaller and more elite fraction of humanity. Once. . \par\pard\plain\hyphpar}{ And what happens a month or two (or a day or two) after that? I have only analo gies to point to: The rise of humankind. Now. but he does not pursue its most d isturbing consequences. 2. (There is the farfetched possi bility that we could make a human equivalent out of less powerful hardware --if we were willing to give up speed. they could put such fantasi es millions of years in the future. Perhaps it was the science-fiction writers who felt the first concrete imp act. sadly. I think I'd be more comfortable if I were regar ding these transcendental events from one thousand years' remove . The dilemma felt by science-fi ction writers will be perceived in other creative endeavors. . within the twentieth century.hines is one of these intellectual activities. \par\pard\plain\hyphpar}{ Through the sixties and seventies and eighties." and the intelligence of man would be left far behind. Now they saw that their most diligent extrap olations resulted in the unknowable . if we were willing to settle for an artificial being that was literally slow. It is more probable than not that. cad/cam) that release us from most low-level drudgery. . . Can the Singularity Be Avoided? \par\pard\plain\hyphpar}{ Well. . we will see the predictions of true technological unemployment finally come true. \par\pard\plain\hyphpar}{ Good has captured the essence of the runaway. But it's much more likely that devising the soft ware will be a tricky process. We will be in the Posthuman era. After all. recognition of the cataclysm sp read. \par\pard\p lain\hyphpar}{ And what of the arrival of the Singularity itself? What can be said of its actu al appearance? Since it involves an intellectual runaway. or chimpanzees. Any intelligent machine of the sort he describes would n ot be humankind's "tool" --any more than humans are the tools of rabbits. Anothe r symptom of progress toward the Singularity: ideas themselves should spread eve r faster. provi ded that the machine is docile enough to tell us how to keep it under control. soon. galactic empires might hav e seemed a Posthuman domain. . . There are the widely respected arguments of Penrose3 and Searle4 against the practicality of machine sapience. instead o f twenty. What abou t the coming decades. After all. maybe it won't happen at all: sometimes I try to imagine the symptoms we should expect to see if the Singularity is not to develop. even interplanetary ones are.) We will see a utomation replacing higher-and higher-level jobs. these writers felt an opaque wall across the future. robins . More and more.)\par\pard\plain\hyphpar}{ But as time passes. and even the most radical will quickly become commonplace. the general cri tics of machine sapience will have good press.

(Or. pretty bad. I argue that confinement is intrinsi cally impractical. Even if all the gover nments of the world were to understand the "threat" and be in deadly fear of it. Instead. there is little doubt that over a period of years (your time) you could come up with a way to escape. our present computer hardware might be as much as ten orders of magnitude short of the equipment we carry around in our he ads. just how bad could the Pos thuman era be? Well . we might never see a Singularity." Such a "weakly superhuman " entity would probably burn out in a few weeks of outside time. If so. \par\pard\plain\hyphpar}{ If the Singularity can not be prevented or confined. military. if the Penrose or Searle critique is v alid). .6 who explic itly cites the development of transhuman intelligence as a sufficient condition to break his projections. This is very like the future predic ted by Gunther Stent. Would a thousand years of doggy living add up to any human in sight? Many speculations about superintelligence seem to be based on the weakly superhuman model. It's hard to say precisely what "strong superhumanity" would be like. The physical extinction of the human race is one possibility. If this is true (or for that matter. . as Eric Drexler put it of nanotechnology: given all tha t such technology can do.7 He agrees that superhuman intelligences will be available in the near fu ture. Commercial digital signal processing might be awesome. the participants were not especially supportive of the arguments against machine intelligence. but without the ability to push it further. \par\pard\plain \hyphpar}{ "Strong superhumanity" would be more than cranking up the clock speed on a huma n-equivalent mind. The competitive advantage --economic. progress toward the goal would continue.) Yet physical extinction may not be the scariest possibilit y. However. . it will." As you might guess from the wo rkshop's title. Think of the different ways we relate to animals. even artistic --of every advance in automation is so compelling that f orbidding such things merely assures that someone else will get them first. It would likely be seen as a golden age . I will re turn to this point. there was general agreement that minds c an exist on nonbiological substrates and that algorithms are of central importan ce to the existence of minds. to your masters. The majority of the participants agreed with Hans Moravec's estimate5 that we are te n to forty years away from hardware parity.tigate "How We Will Build a Machine That Thinks. . A Posthuman world would sti ll have plenty of niches where human-equivalent automation would be desirable: e . I believe that our best guesses about the post-Singularity wor ld can be obtained by thinking on the nat ure of strong superhumanity. But Drexler argues that we can confine such transhuman devices so that the ir results can be examined and used safely. in the early '00s we would fin d our hardware performance curves beginning to level off -because of our inabili ty to automate the design work needed to support further hardware improvements. In fact. \par\pard\plain\hyphpar}{ But if the technological Singularity can happen. A minority felt that the largest 1992 compu ters were within three orders of magnitude of the power of the human brain. I think that any rules strict enough to be effective would al so produce a device whose ability was clearly inferior to the unfettered version s (so human competition would favor the development of the more dangerous models ). but the difference appears to be profound. Eric Drexler has provided spectacular insights about how far technical improvement m ay go. perhaps governments would simply decide that they no l onger need citizens. and it would also be an end of progress. giving an analo g appearance even to digital operations. I call this "fast thi nking" form of superintelligence "weak superhumanity. but nothing would ever "wake up" and th ere would never be the intellectual runaway that is the essence of the Singulari ty. And yet there was an other minority who conjectured that the computational competence of single neurons may be far h igher than generally believed. Imagine yourself locked in your home with only limited data a ccess to the outside. We'd end up with some very powerful hardware. there was much debate about the raw hardw are power present in organic brains. \par\pard\plain\hyphpar}{ Another approach to confinement is to build rules into the mind of the created superhuman entity. Imagine running a dog mind a t very high speed. If those masters thought at a rate --say --one million times slower than you.

they are usually imagining an AI project. it could have the positive effect on . a dedication that would put them in a mental hospital in our era. \par \pard\plain\hyphpar}{ Here are some possible projects that take on special significance. they might be the closest things in the new environment to what we call human now.) Some of these human equival ents might be used for nothing more than digital signal processing. the hardest development problems have already b een solved. (A strongly superhuman intelligence would likely be a So ciety of Mind8 with some very competent components. yet they could lead to the Singularity.D. an enormous amount of research has gone into designing computer aids for artists. I'm suggesting t hat we explicitly aim for a greater merging of competence. that we explicitly re cognize the cooperative approach that is possible. Of course. Considering the bizarreness of higher-dimensional hillclimbing problems (and the neat algorithms that have been devised for their solution). human and good computer workstation (even an offnet workstation) could probably max any written intelligence test in existence. Hum an/computer symbiosis in art: Combine the graphic generation capability of moder n machines and the esthetic sensibility of humans. Eve n the largest avalanche is triggered by small things. \par\pard\plai n\hyphpar}{ Note that I am not proposing that AI research be ignored. it may not be clear what t he right guiding nudge really is: 3. but which serve to advance us toward the Singularity along the IA path. And yet: we are the initiators. AI advances will ofte n have applications in IA. t hat its coming is an inevitable consequence of humans' natural competitiveness a nd the possibilities inherent in technology. Other Paths to the Singularity\par\pard\pla in\hyphpar}{ When people speak of creating superhumanly intelligent beings. to make things happen in ways that are less inimical than others. in some sense we have achieved an increase over natural intelligence. I am suggesting that we recognize tha t in network and interface research there is something as profound (and potentia lly wild) as artificial intelligence. and design programs and in terfaces that take advantage of humans' intuition and available computer hardwar e. yet with a onesidedness. in most cases not even recognized for what it is by its developers. \par\pard\plain\hyphpar}{ And it's very likely that IA is a much easier road to the achievement of superh umanity than pure AI. there are other paths to superhumanity. And there is at least conjectural precedent for this approach. But every time our ability to access information and to communicate it to others is improved. the team of a Ph.mbedded systems in autonomous devices. We have the freedom to est ablish initial conditions. IA is proceeding very na turally. Cairns-Smith9 has speculat ed that biological life may have begun as an adjunct to still more primitive lif e based on crystalline growth. Of course (as with starting avalanches). I call thi s contrasting approach Intelligence Amplification (IA). Lynn Margulis (in10 and elsewhere) has made stron g arguments that mutualism is a great driving force in evolution. But as I noted at the beginning of this article. given the IA point of view:\par\pard\plain\hyphpar}{ Human/computer team automation: Take problems that are normally considered for purely machine solution (like hillclimbing problems). Even now. Though none of these creatures mi ght be flesh-an d-blood humans. and vice versa. Karl Sims has done wonderful work in this direction. we may see projects tha t are not as directly applicable as conventional interface and network design wo rk. With that insight. In humans.11\par\pard\plain\hyphpar}{ Human/computer teams at chess tournaments: We already have programs that can pl ay better than almost all humans. self-aware daemons in the lower functioni ng of larger sentients. But how much work has been done on how this po wer could be used by a human. some very intere sting displays and control tools could be provided to the human team member. to get something even better? If such teams were a llowed in at least some chess tournaments. Others might be very humanlike. Computer networks and human-computer interfac es seem more mundane than AI. Building up from within ourselves ought to be easier than figuring o ut what we really are and then building machines that are all of that. I have argued above that we cannot prevent the Singularity.

. This is generally the area of "groupware". As connectivity. The power and influence of the Inte rnet are vastly underestimated. we are seeing something like Lynn Margulis' vision of the biosphere as data processor recapitulated. and computer speed all increase. Instead of simply trying to model and understand b iological life with computers. in the long run. this suggestion's goal might be to invent a "Rules of Order" for such combination operations. The very anarchy of the worldwide net's developm ent is evidence of its potential. we need to know the fine-scale architecture of vision. This is a form of IA. For example. \par\pard\plain\hyphpar} { Plugging into the optic trunk has the potential for bandwidths of 1\par\pard\pl ain\hyphpar}{ Mbit/second or so. archive size. \p ar\pard\plain\hyphpar}{ Local area nets to make human teams more effective than their component members . In fact. \par\pard\plain\hyphpar}{ Direct links into brains seem feasible. \par\pard\plain\hyphp ar}{ In one sense. A popular research/product area in r ecent years has been decision support systems. but at a million times greater speed and with millions of humanly intelligent agents (ourselves). but giving developing brains access to complex simulated neura l structures might. A longtime dream of science fiction has been direct brain-to-computer interfaces. But for this. the problem becomes vastly more intractable. or for the feature s we don't understand well enough yet to implement in hardware. The above ex amples illustrate research that can be done within the context of contemporary c omputer science departments. produce animals with additional sense paths and interesting intellectual abilities. sitting in front of a computer. concr ete work is being done in this area: Limb prosthetics is a topic of direct comme rcial applicability. much of the work in artificial intelligence and neural nets would benefit from a closer con nection with biological life. the actual brain neuron targets might not have to be precisel y selected. Even 100 bits per second would be of great use to stroke victims who would otherwise be confined to menu-driven interfaces. For instance. Alas.)\par\pard\plain\hyphpar}{ More symmetrical decision support systems. There are other paradigms. group focus might be more easily main tained than in classical meetings. Interfaces that allow computer and network access without requiring th e human to be tied to one spot. As much as the program giving the user information. I had hoped that this discussion of IA would yield some clearly safer approaches to the Singularity (after all. If we want our high-bandwidth connection to add to the paths already prese nt in the brain. the change in viewpoint here would be to regard the group activity as a combination organism. as the embryo developed. there must be the idea of the user giving the program guidance. That suggests:\par\pard\plain\hyphpar}{ Animal embryo experiments. bandwidth. This is an exciti ng near-term step toward direct communication. research could be directed toward the creation of composite systems that rely on biological life for guidance. about all I am sure of i s that these proposals should be considered. Of all the items on the list. Just sticking a gr id of high-bandwidth receivers into a brain certainly won't do it. But suppose t hat the high-bandwidth grid were present as the brain structure was setting up. but may be too focused on systems that are oracular. progress in this is proceeding the fastest. Individual members' \par\pard\plain\hyphpar}{ expertise could be isolated from ego issues so that the contribution of differe nt members is focused on the team project. I wouldn't expect any IA success in the first years of such research.IA research that allowing computers in tournaments had for the corresponding nic he in AI. and we need to place an enormous web of electrodes with exquisite preci sion. IA allo ws our participation in a kind of transcendence). \par\p ard\plain\hyphpar}{ The Internet as a combination human/machine tool. that they may give us more options. if the bit rate is low: given human lea rning flexibility. And of course shared databases could be used much more conveniently than in conventional committee operations. (This aspect of IA fits so well with known economic advantages that lots of effort is already be ing spent on it. Nerve-to-silicon transducers can be made.

the vision fits many of our happiest dreams: a time unending. one where l osers take on the winners' tricks and are coopted into the winners' enterprises. much of what we value (knowledge. To live indefinitely long. perhaps there are rules for distinguishing self from others on the basis of band width of connection. This "problem" about immo rtality comes up in much more direct ways.. I think certain notions of e thics would apply in such an era. Intelligence Amplification undercuts our concept of ego from another direction. when self-awareness can grow or shrink to fit the nature of the problems und er consideration? These are essential features of strong superhumanity and the S ingularity. w here we can truly know one another and understand the deepest mysteries. Thinking about them. That frame is based on the idea of isolated. including ones far higher than sp eech or written messages. For th ose who remained unaltered. The post-Singularity world will involve extremely high-bandwidth net working. the Cairns-Smith or Lynn Margulis notion of new life gro wing incrementally out of the old must still be valid. but that it contradicts o ur most deeply held notions of being. \par\pa rd\plain\hyphpar}{ In fact. thought) need never be lost.But as for safety --some of the suggestions are a little scary on their face. what fellow-feeling can it have with the soul that it was originally? \par\pa rd\plain\hyphpar}{ The later being would be everything the original was. \par\pard\plain\hyphpar}{ 4. it's a lot like the worst-case scenario I imagined earlier. Research into IA and high-bandwidth communicat ions should improve this understanding. the mind itself must grow . I think a closer look at the notion of str ong superhumanity can show why that is. . \par\pard\plain\hyphpar}{ But in this brightest and kindest world. Ulam. Strong Superhumanity and the Best We Can Ask For Suppose we could tailor the Singularity. The problem is not simply that the Singularity represents the passing of humankind from center stage." Bulletin of the American Mathematic . Immorta lity (or at least a lifetime as long as we can make the universe survive) would be achievable. immutable m inds connected by tenuous. the philosophical problems themselves become intimidating. I A for individual humans creates a rather sinister elite. From an other angle. A central feature of strongly superhuman entities will likely be their ability to communicate at variable bandwidths."12 \u166?\par\pard\plain\h yphpar}{ 1. "God is what mind becomes when it has passed beyond the scale of our comprehension. Much of that deadliness may not be necessary in today's world. and when it becomes great enough. And while mind and self will be vastly more labile than in the past. memory. "Tribute to John von Neumann. after a few thousand years it would look more like a repeating tape loop than a person. We humans have millions of years of evolutionary baggage that makes us regard competition in a deadly l ight. Suppose we could attain our most extravagant hopes. \par\pard\plain\hyphpar}{ . the goal would be benign treatment (perhaps even giv ing the stay-behinds the appearance of being masters of godlike slaves). What then woul d we ask for? That humans themselves would become their own successors. but vastly more. and looks back . Ye t now the notion of self-awareness is under attack from the artificial intellige nce people. A creature that was built de novo might possibly be a much more benign entity t han one based on fang and talon. . \par\pard\plain\hyphpar}{ From one angle. low-bandwith links. It coul d be a golden age that also involved progress (leaping Stent's barrier). A mind that stays at the same capacity cannot live forever. And so e ven for the individual. The notion of ego and self-awareness has been the bedrock of the hardheaded rationalism of the last few centuries. . I think Freeman Dyson has it right when he says. What happens when pieces of ego can be copied and merg ed. one begins to feel how essentially strange and different the Posthuman era will be --no matter how cleverly and benignly it is brought to be. that wha tever injustice occurred would be tempered by our knowledge of our roots. S. I think the new era is simply too different to fit into the classical frame of good and evil. I see just the glimmerings of this now. But the post-Singularity world do es fit with the larger tradition of change and cooperation that started long ago (perhaps even before the rise of biological life).

Poul.. ed. Niven.. Run!". "How We Will Build a Machine That Thinks. writing as Olof Johanneson. Ber keley Books. 1992. Frank. 3. "Runaroun d. 64. 8-4 0. 1986. vol. S.. 1983. Tipler. The Coming of the Gol den Age: A View of the End of Progress. Anchor Press/Doubleday.February 1987. April 1967. For information regarding reprints. Alt and Morris Rubinoff. Marvin. January 1983. May 1958. 1961. 1981. "Towards an Artificial Brain. \par\pard\plain\hyphpar}{ Dyson.al Society. et al. Karl. private communic ation. August 24-26. The Anthropic Cosmological Principle. 1989. et al." Review of Modern Phy sics. 3. Isaac. Searle. 1989. Roger. "First Word. Paris. 39. 1991. Kovacs. Berkeley Books. Eric. vol 6. John R. Cairns-Smith." Astounding Science Fiction. Unauthorized copying of this information is strictly forbidden. Vernor. 5. 51. "Minds." Binary Star Number 5. Mind Child ren. Simon and Schuster. and Programs. Michael.. Charles. "Physics and Biology in an Open Universe. A. A. 12. Analog. 6. Vinge. Academic Press." in The Behavioral and Brain Sciences. Harvard University Press. vol. F ranz L. 893-902." in Emergent Computation. Society of Mind. "Regenera tion Microelectrode Array for Peripheral Nerve Recording and Stimulation.. Summit Books. Kurt. Lynn and Dorian Sagan. Penrose. 1969. Harper & Row. December 19 91. 2. MIT Press. pp. Oxfo rd University Press. Other So urces\par\pard\plain\hyphpar}{ Alfvin. "Computational Connectionism within Neurons: a Model of C ytoskeletal Automata Subserving Neural Networks. 94. Seven Clues to the Origin of Life. \par\pard\plain\hyphpar}{ Rasmussen. "Interacti ve Evolution of Dynamical Systems. \par\ pard\plain\hyphpar}{ 4. \par\pard\plain\hy phpar}{ \par\pard\plain\hyphpar}{\par\pard\hyphpar }{\page } } . Hans. Asimov. Swanwick Michael. 17 5-218. Vinge. Cambridge University Press. 1-49. John D. Dyson. Olaf." Analog Science Fiction-Science Fact. 8-36. \par\pard\plain\hyphpar}{ Herbert. no. Dell. Stent. \par \pard\plain\hyphpar}{ 10. 447-460. 10." Omni. Award Books. \pa r\pard\plain\hyphpar}{ Conrad. G. Vernor. \par\pard\plain \hyphpar}{ 8. Brains. "True Names. 31-88. Thearling. "Speculations Co ncerning the First Ultraintelligent Machine. 11. \par\pard\plain\hyphpar}{ Barrow. 1986.. Cambridge University Press. Vernor. 1985. T. Platt. "Kings Who Die.. Engines of Creation. \par\pard\plain\hyphpar}{\s1 \afs32 {\b \par\pard\plain\hyphpar}\par\pard\plain\hyphpar}{ -----------------------------------------------------------The contents of this file are copyright 1993 by the publisher in whose directory this file appeared. Oxford University Press.. Freeman. 1965..Com\pa r\pard\plain\hyphpar}{ -----------------------------------------------------------. Larry." IEEE Transactions on Biomedical Engineering. The End of Man?. 1980. J. Microcosmoss: Four Billion Years of Evolut ion From Our Microbial Ancestors. March 1966. Steph anie Forrest. no. 1985. 1969. \par\pard\plain\hyphpar}{ Bear. Good. "Bookworm. Vacuum Flowers." BioSystems. K. Infinite in All Directions. Please read the general notice at the top menu of the Gopher Server for the Electronic Newsstan d. " The Ethics of Madness. Dune. Moravec. Hannes. \par\pard\pl ain\hyphpar}{ 3. MIT Press. March 1962. The Starmaker." If. please send mail to REPRINTS@Enews." Thinking Machines Corporation. Stapledon. Sims. 1986. et al. 1988. I. 1988. The Natural History Press. 23. "Blood Music. 1979.. 7. March 1942. serialized in Isaac Asimov' s Science Fiction Magazine. and Frank J. Vinge. Gunther S. December 1986 . Technical Rep ort Series (published in Toward a Practice of Autonomous Systems: Proceedings of the First European Conference on Artificial Life. 82-108. 9. 428-449. 1985. Minsky." a workshop at Thinking Machines Corporati on. Drex ler. The Emperor's New Mind. \par\pard\plain\hyphpar}{ Anderson. vol. Margulis. Greg." in Advances in Computers. G." If. June. eds. 9. vol. Freeman. vol.