©2000 by Eliezer S. Yudkowsky.
All rights reserved.
Please do not quote this material, in whole or in part, without permission of the author.
Describes Eliezer as of August 31st, 2000.

NOTE: Update:  As of June 6th, 2000, I have moved to Atlanta.  I am now a Research Fellow of the Singularity Institute for Artificial Intelligence.

NOTE: This should not be the first Low Beyond page you read.  I assume you're here because you read one of my pages and started wondering about the author.  (If not, try "Staring into the Singularity", or "Coding a Transhuman AI", or "Algernon's Law", or "The Plan to Singularity".)  I'm assuming that you are already interested in the questions, and I can concentrate on the answers.  If not, please don't start here.

NOTE: I don't want people quoting sections of this page out of context, so, as copyright holder, I specifically deny permission to quote this page in whole or in part.  If you want to reproduce so much as a sentence, then please just ask me.  Likewise, please do not mirror or duplicate this page.

1: My life so far

1.1: Birth

On September 11, 1979, I was born.

1.2: First memories

I remember brief flashes:  Hiding in an (empty) laundry hamper and closing the lid; looking at blue designs on beige wallpaper while running through an airplane; listening to the "Star Wars" theme during nap time in pre-nursery.

I remember getting a new squirtgun, of translucent blue plastic, which squirted better than any squirtgun I'd had so far.  And within around an hour, for some reason or other - I don't think it was related to abuse of the squirtgun, but I'm not sure - my mother decided to take it away.  So I rushed out the back door and threw it off the third-floor balcony, and listened to it shatter on the concrete below.

I knew, as I threw it, that Mom would think it had been a childish act - "Cutting off your nose to spite your face," was what my parents used to say.  But that's not what it was.  Grownups who have forgotten their childhood may lend a toy to a particular child, and not lend it to anyone else, but they don't credit children with enough "agenthood" to really own something.  Rather than let my parents say, in essence, that this toy - which had been given to me, and which I really enjoyed - was theirs to lend or withhold, I threw the toy off the balcony and accepted the consequences of not being able to play with it any more.  Not because I didn't want them to have the toy, but because it was a wonderful toy, the best squirtgun I'd ever had; that's why I cared enough that I threw it off the balcony rather than let it be something for my parents to give or take away.  When I heard the squirtgun shatter, I didn't feel angry, or regretful, or look-what-you-made-me-do.  I felt complete.  I'd done what I'd set out to do, and I was willing to accept losing a squirtgun in order to do it.  Children can also act on principle.

I remember building a ski slope out of wooden blocks; it wasn't the greatest invention of my early years, but it was certainly the most popular.  I remember taking a turn through the structure, my classmates gathering round to look; I remember giving permission for my classmates to use it,  telling them to form an orderly line for using it, skipping my own turn on the grounds that I'd already had one, and watching as my classmates skied up and down my sloping wooden structure.

It wasn't exactly a life-changing event, but it's my first memory of stepping back from the group and playing guardian.  For a long time, I would have said this was a happy memory.  I would now say that no, watching from the sidelines while other people have fun isn't a source of happiness.  Even so, it seems to be a choice that I keep making.

I remember being called a crybaby.  Crybaby is a pejorative term, but taken as a description, it was accurate; one of my earliest memories is crying because someone else wouldn't let me put away a toy telephone during cleanup time.  And at the age of five, or so - possibly before - I remember reading "Childcraft" books with my grandmother.  I had no patience for "Places to Know" or "Poems and Rhymes", but I read "World in Space", "The Green Kingdom", "How Things Work" and "Mathemagic" repeatedly.  And I understood and retained it; another of my earliest memories (1) is learning about leverage and remembering what a "fulcrum" was the next day.  The person reading me the book, who had already forgotten, was very impressed by this (2).  (And in fact, to this very day, I still know what a fulcrum is.)  Two early signs of the two great themes.

1.3: School

By the time I got to kindergarten, and received my first lessons in reading and writing, I had already learned everything they were teaching.  This set a pattern that remained constant throughout the rest of my school years.

By first grade, I had set another pattern; there were one or two kids that I would talk to, and I didn't interact much with the rest of the class.  It's not that I was the class nerd; even that far back - far enough back that I was still struggling with the "crybaby" label - I was the class genius and everyone knew it.  Even being picked on is a social position, but I wasn't part of the social mesh.  Sometimes the social mesh would reach out and touch me, generally to make me unhappy; I reacted by walking away, returning fire, or pretending to be happy, depending on circumstances.  It would be a mistake to say that I liked being outside the group, and equally a mistake to say that I wanted to be inside it.  My life didn't revolve around the other kids; it revolved around one or two friends, around my parents, and mostly around the books I was reading.

I certainly wasn't "picked-on".  The class had two or three bullies.  Once, two of them (3) tried to take away my basketball during recess.  (And when I say my basketball, I mean one that I personally owned and had brought from home, not just one that I happened to be holding at the time - I can still remember the feeling of indignation.)  Even back then, I wasn't very athletic, and it was two against one.  So I held the basketball behind my back with one hand, punched one of them, kicked the other one, and walked away.

I knew what a "solar plexus" was, and they didn't.

I suppose I'm fortunate that this issue was settled far enough back that one piece of real knowledge could make that much of a difference.  Anyway, I didn't have the miserable, low-social-status youth that a lot of bright kids have, and I think that one incident is one of the major reasons why.  I was teased, but by individuals or small groups - not by the class.  Perhaps if I'd participated in the social games - if I'd had no choice but to participate, or if I'd wanted to participate - I'd have had a miserable youth.  But when the social mesh reached out to engulf me, I could often either walk away, or make it leave me alone.

In second grade, I was shocked to learn that my math teacher didn't know what a logarithm was.  (Not to give you the wrong impression, at the time, I didn't know what an "exponent" was.  My parents called them logarithms, so that's what they were.)  I permanently lost all respect for my teachers, and for the entire institution of school, and started pleading to be taken out.  My parents told me that I had to go to school, even if I wasn't learning, to learn how to interact with the other kids.  I said that if that was the case, they should send me to a specialized institution for learning how to interact with other kids, because I certainly wasn't learning any social skills in school.  (4).  In retrospect, I would still have to say that I was right about this (5).

It was during second grade that I bit a teacher.  But it was just the one time.

In third grade, things continued onward in the same old pattern.  I didn't interact much with the other kids, but I was the class genius, and I was always in trouble with the school authorities, so nobody picked on me, except for individual-to-individual teasing.

I continued reading adult-level books.  It was in fourth grade, at the age of nine, that I read Richard Feynman's QED.  (And, may I note, understood most of it.)  At the time, I thought I would grow up to be a physicist.  Dad was a physicist; I'd liked reading QED and the physics books I'd gotten hold of before that; I liked knowing what a quark was (one of only two fourth graders to do so); therefore, my default assumption was that I would be a physicist.

It was also at the age of nine that I tried to write a science-fiction novel, producing sixty pages of appalling garbage (6) which I recently managed to find on an old floppy disk.  (My other computer files - all my old school papers, and so on - were lost when my father accidentally destroyed them.  No backups.  I can still feel the hole.)

In fifth grade, my parents first tried out the concept of accelerated classes; I took a sixth grade math course.  It wasn't any different from previous math courses, in the sense of being yet another review of knowledge I'd had since first grade, but I supposed it was better than nothing.  So, next year, my parents agreed to let me skip sixth grade and go directly to seventh.  I was enthusiastic about the idea because it'd get me out a year earlier; also, I had three friends in that grade.

In seventh grade, the other kids picked on me, which was a new experience.  They didn't know me, and they didn't know better, and - unlike kids my own age - I couldn't hold my own against them in a fight.  But for the first few months, I managed.

Also in seventh grade, I learned about the concept of rising and falling tension, and realized what was wrong with my fourth-grade novel; it didn't have a plot.  This is literally the only significant thing I learned in school, from pre-nursery to eighth grade.  Was it worth ten years?  No.

1.4: Great Mambo Chicken

At the age of eleven, I read a book called Great Mambo Chicken and the Transhuman Condition.  My uncle (grand-uncle, actually) had borrowed it from the library, presumably because it looked like something I might enjoy.  I enjoyed it a great deal.  I won't say that I resolved to embark upon a career in one of these areas, because I don't remember resolving anything.  I just remember knowing that this was what was important in life, and that my career was going to touch on one of those areas.

Thanks to a steady diet of science fiction, and J.E. Pournelle's A Step Farther Out, the idea of space travel and such was already firmly planted in my mind - and beyond that, the idea that there was a better world than this one.  In the Step Farther Out days, this better world consisted of the vast wealth locked up in the asteroid belt, the spaceship designs infinitely preferable to "disintegrating totem poles" - NERVA and so on - that dumb ol' NASA wouldn't use, and so on.  And I always imagined myself as not only living to see that future, but helping to bring it about - being the theoretical physicist who discovered stardrive, or the person who brought the first trillion-dollar asteroid home.

Great Mambo Chicken added nanotech and the other ultratechnologies to the list of technophilic preoccupations, and upped by several orders of magnitude the "better world" stakes.

1.5: The 1991 Midwest Talent Search

In seventh grade, at age eleven, I signed up for the Midwest Talent Search, run by the Center for Talent Development.  CTD administered the SAT to 28,000 sixth, seventh, and eighth graders (7) across, you guessed it, the Midwest (8).

I obtained a couple of SAT preparation books - one targeted specifically on Math, and one targeted on the whole SAT (Math and Verbal).  I took a few practice tests from the Math book, and with each additional test, my scores went down.  I got a 570, then a 530, then a 460 (9).  "Huh?" I said to myself.

While scoring my third test, I noticed that sometimes I'd answered correctly, then erased it.  So that was the problem:  I was second-guessing myself.  I'd come up with the correct answer, question it, and replace it with a wrong answer.  Each time my test scores went down, I would double-check myself even harder on the next test.  I wasn't trusting my intuitions.

So I took another practice test, this time resolving to, as Ben Kenobi would say, "act on instinct".  (That actual phrase, in Ben's voice, ran through my head.)  (10).  I got a 640 Math.  The lesson I learned was to trust my intuitions, because my intuitions are always right - probably one of the most important lessons of my entire life.

On my actual SAT, I got a 670 Verbal and a 740 Math.  The Midwest Talent Search informed me that this had placed second Verbal, third Math, and second Combined, for the seventh grade, for the Midwest.  Their statistics said I was at the 99.9998th percentile.  It wasn't until years later that I realized their stats were worthless because I'd skipped a grade, and to this day, I still don't know what percentile I'm really in.

This was the first real sign that I was not only bright but waayy out of the ordinary.  But it had less emotional impact than you might think.  My parents warned me not to become arrogant.  The cultural wisdom embodied in all the science fiction I'd ever read warned me not to become arrogant.  Thinking that test scores made you "better" was portrayed as a mistake, and I believed in that portrayal.  (11).  And it's a remarkable thing, but when a mistake is depicted in great detail, and your parents are watching you to ensure that you don't fall into it, and you believe it's a mistake, then, by golly, you can actually avoid making the mistake.

I once read, in a history of Microsoft, that Bill Gates placed ninth in a statewide math competition during elementary school, and that this "settled his place in the Gates household" (12).  No such event took place in our household.  The common feeling - including for me - was that these were excellent test scores, but they were not sufficiently out of the ordinary to change anything.  (13).

1.6: Falling out of school

At the end of seventh grade (14), when I was around eleven and a half, I suddenly lost the ability to handle school.  I stopped doing my homework.  Instead of going to classes, I would sit in the school office, crying, until my mother picked me up.  I am told that I made it through eighth grade and graduation, but I remember little or nothing of it.  I don't recall it as a period of intense misery, except when I was actually in the classrooms (15); I do recall it as a period when I spent a lot of time crying.

1.7: "I can't do anything."

When school was over, high school was pretty much out of the question.  There was no chance I'd be able to handle it.

"I can't do anything."  That's the phrase I used then.  To describe it in terms that might be tainted by later interpretation (16), I couldn't exert my will.  I had no mental energy.

Trying to do anything would exhaust me, especially if it was something I didn't want to do.  You can imagine how well that went over with two people trying to apply normal parenting heuristics to the situation.  I can imagine how well it's going over with my average reader right now.  But a spoiled brat exhibits defiance, and I couldn't even do that.  I could argue, rationally, with my parents.  I would break down in tears if the conversation got stressful, but I could keep arguing rationally (17).

My parents would have a brand new idea for what I should do next.  I would see the disaster coming, and try to convince them of it.  They would not be convinced.  They would keep pushing me, and I would go along with it - not because I agreed, but because I couldn't stand up to them.  Because my will was broken.  Because I was only thirteen.  (And fourteen.  And fifteen.  And sixteen.)  I would try, and I would run into a problem, and I would break.  And my parents would have a new bright idea, and the pattern would repeat.  Over and over.

My parents were well-meaning people, totally out of their depth and refusing to realize it.  They relied on their memories of being bright kids themselves, with no conception of the gap.  They were immune to logic because, in their memories of being smarter than their parents, they could prove anything with logic and it didn't mean anything.  So they could safely ignore anything I said, any argument I made, because it was just the sophistry of a bright child.  They were parents now.  It was their turn to be in charge of the situation, and nobody was going to take that away.

It was a classical, predictable, entirely natural mistake, and I don't blame them for it.  It was the Bayesian Probability Theorem that got them.  In the vast majority of situations where the parent thinks they know what's best for a child, they do.  But sometimes the theory breaks down, and then someone has to pay the piper.

I never hated my parents.  They said they loved me, and the cultural wisdom of the science-fictional literature said your parents genuinely love you no matter what you disagree about.  The books I read showed parents who cared about their children, and the foolish, unwise children rejecting them out of anger.  I believed in that scenario, and I refused to give in to it.  People can love you and care about you and want what's best for you and still use their power to screw up your life and put you through hell.  Behold the foundations of my morality.  (19).  Sometimes, awful things happen and there isn't anyone you can rationally blame for it.  (20).

1.8: True Names and Other Dangers

NOTE: My memory of the time when something happened has always been very sketchy, and I don't really recall the relative order of these events, so the given order should be considered unreliable.  Nonetheless, this collection of events - "True Names", "Algernon's Law", and "Breaking point" - would have started happening at late fifteen or early sixteen.

As you've probably already read in one of my pages, I recall the exact point at which I became a Singularitarian.  It was on reading page 47 of True Names and Other Dangers by Vernor Vinge.

"Here I had tried a straightforward extrapolation of technology, and found myself precipitated over an abyss.  It's a problem we face every time we consider the creation of intelligences greater than our own.  When this happens, human history will have reached a kind of singularity - a place where extrapolation breaks down and new models must be applied - and the world will pass beyond our understanding."
My emotions at that moment are hard to describe; not fanaticism, or enthusiasm, just a vast feeling of "Yep.  He's right."  I knew, in the moment I read that sentence, that this was how I would be spending the rest of my life.  It was just so obvious.

I've been a Singularitarian ever since.

1.9: Algernon's Law

Over the course of years, I'd thought of dozens of explanations for why I couldn't do anything, tried dozens of schemes for fixing it.  None of them worked.

I get the impression that the usual rule is that your mind appears to work however you expect it to work.  I can even remember falling victim to this circular form of self-delusion, back when I was shuffling explanations - back before I learned to consciously observe the effect of expectation on perception.  But regardless of what theory I came up with, and what solution I tried, I still didn't have any mental energy.  The placebo effect was not a match for my problem.

I remember the exact moment at which light broke - I'll describe my initial feelings in a bit more detail below.  But first, the theory in its present-day form.

There's a single emotional tone - an emotional tone is a modular component of the emotional symphonies we have English words for - common to sorrow, despair, and frustration.  The tone is invoked by an effort failing to produce the expected reward ("frustration"), or by the anticipation of something going wrong ("despair"), or by watching something go wrong ("sorrow").  The message of this tone can be summarized as:  "This isn't working.  Stop what you're doing, try to figure out what you're doing wrong, and try something else."  The cognitive methods activated by this tone (21) include what I would now call "causal analysis", "combinatorial design", and "reflectivity".  The motivational effect of the tone includes, of course, low mental energy.

To get an idea of what this tone feels like, tilt your head back and try to scream, inaudibly, at the pitch bats use; then, add the burning sensation you get at the back of your throat when you're about to cry; the result is the tactile aspect of the tone.

That catch in the throat is always with me.  It is present when I get up in the morning, when I go to sleep at night, and at every moment in between.  Such emotions often have specific neurological substrate, and it is known that neurological perturbations can alter or obliterate both entire emotions and specific facets of emotions.  (22).  The catch in the throat, and the low mental energy that goes with it, and the cognitive methods it invokes, are constantly present in my mind.  These characteristics are "nailed down", present at all times and regardless of external conditions.

Thus a single cause neatly accounts for both my SAT scores and my Great Depression, with the available evidence suggesting it goes back to birth or earlier (23).  I have no idea whether this perturbation, this "neurohack", is genetic or prenatal or the result of some disease in infancy, but in retrospect, it's clear that it goes back as far as I - or my parents - can remember.

For more on the general theory of neurohacking, the tradeoffs in Specialization, and my own particular neurology, see Algernon's Law.

I remember the exact moment at which light broke.  I even remember where I was sitting (in a train).  I was, once again, pondering the question of why I didn't have any mental energy, and I tried thinking about the occasions when I did find mental energy.  It occurred to me that when I started a new project, my energy level went up briefly before crashing.  Maybe, I thought, energy was produced by new ideas.  And that's when the light went on.  "Maybe both the genius and the energy deficit were produced by overloading a single force, the force that resists thoughts moving repeatedly in the same channel."  (24).  And then I thought:  "Maybe that's why my genius isn't an evolutionary advantage."

The details were all wrong, of course.  That's not how the brain works.  But the basic idea was right, and within a short time (25), the Algernic hypothesis had evolved into essentially its current form.

The moment at which I started "putting myself back together" would be difficult to pinpoint exactly.  I could trace it to the beginnings of mental discipline - the moment I caught the feeling of refusing to look at something in the corner of my mind (about which more later).  I could trace it to the moment I first read about evolutionary psychology (about which more later).  I could trace it to a certain very unpleasant moment (about which more shortly).  But I think I'll always trace my dawn to the instant when I understood, when I was no longer fighting blind, when I could consciously invent workarounds - workarounds that actually worked - based on my knowledge of what was happening.  Unfortunately, the process was far from instantaneous.

1.10: The breaking point

"I would try, and I would run into a problem, and I would break.  And my parents would have a new bright idea, and the pattern would repeat.  Over and over."  Does this sound unpleasant?  It was.  But, once I broke, once I ran out of willpower, it was usually over in a couple of weeks, and then it would take another couple of months before my parents came up with something new.  And that was enough to keep me sane, if not in very good condition.

Once, they kept me in the situation after it started going bad.  That was probably the single most unpleasant time in my life.  I don't really want to talk about it, certainly not in concrete terms, and certainly not on the Web.  But imagine, if you will, a rat, in a cage.  Each time the rat presses a lever, the rat gets an electric shock.  We'll suppose that this rat has parents who keep pushing at it to press the lever, and that this rat has a quirk of neurology that amplifies the emotion of frustration, in particular the this-isn't-working-try-something-else component.  What you get, after a couple of years, is what I would describe as "positive feedback in negative reinforcement".  It's not just the intrinsic frustration that's the problem; after three years of failure and negative reinforcement, my eternal "this-isn't-working-try-something-else" signal had amplified to the point that it was a separate and distinct electrical shock.  And then the positive feedback cycle got started.

I was sixteen, my will was broken, and I didn't have enough experience to visualize all the options an adult would perceive - leave home and stay with a relative, leave home and call a charity, et cetera.  I didn't see them.  With my will broken, there was no way I could get a job; the world outside my home was an opaque lump, a lethal dead end that could only end with my lying dead in a ditch.  I couldn't simply walk out of the cage, disengage from the situation; my parents would have threatened to throw me out of the house (26), which was a threat I took seriously, and I would have broken.  I couldn't maintain a negative course of action, refuse to do something.  What I could do was adopt a positive course of action, even if it was an act of absolute desperation; that was a resolution even my broken will could maintain in the face of parental opposition.  And my parents, for once, took me seriously - correctly so - and let me out of the cage.  I wouldn't say that I won.  A situation like that doesn't have winners.  But they lost.

That was when I grew up.  Or to be more precise, I grew up two weeks later.

There's a Dilbert cartoon in which Dilbert thinks that he feels like he's been faking everything since sixth grade, and he wonders whether anyone else feels that way.  One morning, two weeks after the confrontation, I woke up knowing that I'd never feel that way again.  That's my current theory of maturity:  You deal with intense emotional stress and a life-threatening situation (27), and two weeks later, you stop feeling like you're faking it.

If I was someone else, I would say that this is because, once you've faced that sort of threat, you become someone who can deal with that sort of threat.  You can take things seriously, visualize the consequences without flinching away; you know you're not faking it.

But I don't think it's a matter of perception, or belief.  My experience as a neurohack has taught me to distinguish between hardware and software.  I believed then, and I believe now, that it's probably a neurological thing, some hormonal response.  I know what a belief-based revision of the emotional landscape feels like; this one seemed to take place without conscious thought being part of the process.  Besides, if the feeling of maturity was a software process, it would have shown up right away - when, may I note, it would have been useful - instead of two weeks later.

(If I were someone else, I would probably philosophize that this growing experience demonstrates the necessity of stress in our lives.  Maybe it does, for human neurology.  But if so, that necessity is something we should dispense with.  Anyway...)

1.11: Searching for support

That maturity  inaugurated the next era of my life, in which my parents were content to let me try my own plans.  (Previously, whatever failing idea my parents were currently enamored of was the dominant fact in my life at any given point.)  The basic dilemma, at that point, was how to support myself with high school, much less college, an impossibility.  Working at a nine-to-five job, much less flipping burgers at McDonald's, would have burned far too much mental energy - on a good day, I can match the energy output of an ordinary human, but that level of effort is impossible for more than a week.

1.11.1: Commodities programming

For two years, from late sixteen through late eighteen, I tried writing a commodities-trading program, by request, for a friend.  Eventually I realized that trying to outprogram the stuff already on the market was three years of work for a full team of programmers; I might be able to do the job of a full team, but I couldn't do it in less than three years.  I wasn't willing to spend another year, so the project halted.  I did wind up with a deep understanding of C++; I think probably up to professional standards, maybe a bit beyond.  And yes, that outcome was part of my initial calculation in taking the job.  I won't say that I thought the commodities program was doomed; I wouldn't have started the job if I'd expected to fail.  But I did know that it was a tenuous plan, with a high probability of failing, and I planned to make that outcome acceptable.  I wound up with a lot of highly reusable code, currently gathering dust.  And I invented the basic design patterns that went into Flare.

Why'd I take on a quixotic project, a tenuous gamble like that?  Well, partially because it was there.  It was something to do, something I could show my parents that I was doing.  And it proved to me that, setting my own hours, and armed with knowledge of how my mind worked, I could work on something for two years without breaking.  Part of it was also the immense payoff that a successful trading program would have meant; since childhood, I'd always imagined myself becoming rich first, then funding my own dreams.  I didn't give up that part of myself until - well, see below.

1.11.2: CaTAI

I formally gave up on the commodities-trading program a few months before my nineteenth birthday, after which I spent a month writing the first version of Coding a Transhuman AI.  (The time was deliberately chosen so that, years later, I would be able to say I had done it "when I was eighteen".)  (28).  Writing CaTAI was an amazing learning experience.  I knew a lot more about minds when I finished than when I started.

I do my best thinking into a keyboard.

1.11.3: Learning about venture capital

After I published the "temporary" version of CaTAI (it would be more than a year and a half later before I started work on the "real" version), I was contacted by someone who was interested in building the AI, if the building could be done by myself or a small team.  Ve asked what it would take to implement it.  I answered "Manhattan Project".  (Note:  I have since changed my mind about this.)  However, in the course of our discussions, I noted that I had all these cool design patterns left over from the commodities project, and suggested turning my code into a consulting company.

At the time, I didn't know how the software industry worked.  After doing some research in Extended Long-Term Memory (i.e., the Web), I learned that a brilliant design pattern is worth basically nothing.  And my sponsor had vis own qualms.  So I took the qualms, treated them as requirements, and suggested a different business idea.

I spent the next six months (A) trying to write the initial corporate Webpages, (B) trying to come up with a design for the initial software, and (C) surfing the Web.  I ran the Web looking for possible competition, looking for the software I'd need, learning about what startups are like, learning how to write a business plan, and finally came to the conclusion that the level of funding we were discussing just wasn't practical for putting a real corporation together.  At most, it would have been "seed funding", not the one or two rounds of venture capital that would have been required.  My potential backer backed out at around the same time, due to a change of personal philosophy (ve retired).

I did come out of the experience with a much better idea of how the software industry worked - in fact, with a much better idea of how real life, in general, worked.  I began to realize that it was not really practical to plan on becoming rich and funding my ideas myself.  Founding a startup company requires charisma and tremendous amounts of energy; you can get by without brilliance, but not without the energy.  And there would have been other problems as well (29).  After a while, I admitted to myself that "getting rich and funding everything personally" might be the most emotionally satisfying way to imagine it - the way I'd happened to picture it back in my childhood, when my core dreams were being formed - but it wasn't the fastest and most solid way from point A to point B.  So I let go of that part of myself.

1.11.4: Flare and PtS

After halting work on the ill-fated business concept, I spent the next few months writing 500K of design notes for a programming language called Flare.  Along the way (on April 28th, 1999, according to my notes), I realized that I could map out a complete technological path from Flare to Coding a Transhuman AI 1.0.  I don't often feel elated, but I felt elated then.  It was a long and winding path - an AI is nothing remotely like a programming language - but, for the first time, I could see a specific sequence of manageable efforts, creating the potential for the potential for the potential, that would end up at the Singularity.

That was when I decided to write The Plan to Singularity.  As I recall, I kept working on Flare until around June, spent some time putting together the Singularitarian mailing list, then spent until January 1st, 2000, writing The Plan to Singularity and the Singularitarian Principles.  It took a lot longer than I expected, but I'd say it was worth it, even though major parts of PtS are already obsolete.  (For example, as my understanding of AI increased, I backed away from the idea that it would take an AI industry, and even from the idea of open-sourcing the project.)

1.12: The Singularity Institute

On March 28th, Brian Atkins first expressed interest in starting a Singularity Institute.  On May 19-21, Brian Atkins, Sabine Stoeckel, and I discussed it at the Foresight Spring 2000 Senior Associate Gathering.  On May 21, we decided that I should move to Atlanta and get the Singularity Institute started.  On June 6th, I moved to Atlanta.  On July 27th, the Singularity Institute for Artificial Intelligence, Inc. received its incorporation papers from the Georgia Secretary of State.

I am now a Research Fellow of the Singularity Institute for Artificial Intelligence.  My current plan is to finish up CaTAI 2, after which I should have a design document sufficiently detailed that the Singularity Institute can hire a project team and begin implementation of the first stages.

2: Where do you live?

Atlanta, Georgia
United States of America
Human Space

2.1: Where were you born?

Chicago, Illinois
United States of America
Human Space

3: What do you do for fun?

Believe it or not, I read science fiction, listen to music, and watch television.  Since moving to Atlanta, I have even begun watching movies.

3.1: What are your favorite books?

Take a look at the Bookshelf.

3.2: What kind of music do you like?

Here's a sampler of my favorites:

As you can see, I'm not picky about which genre a song belongs to.  It just has to be a good song.  My philosophy is that true artists and lousy hacks exist in every tradition.

Except for opera and rap, which I can't stand.

(In association with Amazon.com.)

3.3: What's your favorite television show?

Actually, there's only one television show worth watching.  This show is by far superior to every other television show I've ever seen, up to and including Babylon Five.  If you've ever seen this show, you already know what my favorite show is.  Not even the fraction of a doubt has crossed your mind.

Buffy the Vampire Slayer.

3.4: You watch Buffy the Vampire Slayer?

"Eliezer watches Buffy?  That's wonderful!  So he is mortal, after all."

I get that reaction often, always from people who have never seen the show.  Anyone who has seen the show, especially the second season, knows better than to be reassured.

NOTE: In recent times - the fourth season - the quality has gone far downhill, alas.  The comments below should be taken as referring to the first through third seasons.

But just to set a few matters straight:  Buffy is an extremely intelligent and high-quality show.  Magic has rules on Buffy; they're never explained, but they're very clearly there.  All the people behave like real human beings. Buffy is "realistic", in the sense that it depicts exactly what would happen, given the show's premises.  The characters are intelligent and emotionally mature.  Finally, I do not watch Baywatch, and I see no reason why people should assume that I watch Buffy for the same reason they watch Baywatch.  (30).

That said...

"I wish I could be a lot of things for you.  A great student, a star athlete, remotely normal.  I'm not.  But there is something I do that I can do better than anybody else in the world..."
            -- Buffy Summers, to her mother, in Graduation Day 1
Do you even need to ask?

Well, yes, you do.  It'd be a mistake to focus too much on the fact that Buffy Summers is the one girl in all the world with the strength and speed to hunt the vampires, and I'm the one Specialist working on AI.  Remember, I have no emotional need to be special.  I've spent my whole life being special.

We all have secret little ridiculous dreams.  It's a large part of who we are, especially when it comes to determining our favorite television shows.  One of my secret little dreams is being able to punch through walls.  Given the popularity of The Matrix, I imagine it's a fairly common fantasy.  But do you suppose that Buffy Summers, who can punch through walls, has an emotional need to do so?  There's an enormous difference between the emotional makeup of someone with a fantasy and someone with the actuality, and one of things I deeply respect about Buffy is that they get this exactly right.

Buffy Summers is the first person I've ever seen on television who I can sympathize with.  She's enough of an altruist that I can imagine myself as her without flinching, but she also leads something resembling a normal social and emotional life.  That's a combination I've never seen on television before.  For the first time, there's a competently written show, with a character I can sympathize with, who nonetheless has things that are missing (31) from my own existence.

I spend a lot of time looking out through the eyes of superintelligence, a view that tries to eliminate everything that belongs to human minds or evolved minds rather than minds in general.  Watching Buffy enables me to do a "hard sync" with normal life (32), run the brain under its design conditions for a while.

I imagine that most people can sympathize with Buffy because she leads a normal emotional life, and they watch the show because they want the vicarious thrill of trying to save the world.  Take a moment to savor the irony.

4: What's it like inside your head?

I'd describe it, but - well, I think this story has to be told in chronological order.  So, and I beg your forgiveness, it's back to that childhood Eliezer once more.

4.1: Beginnings

When I was young, I believed in psychic powers on the basis of reports that I later learned were outright lies.  (33).  Some good came of it, however; it was while testing myself for psychic powers that I first started paying attention to my awareness - or, as I would now put it, paying conscious attention to self-perception, the reflective intuitions.  (34).  Anyway, I would practice, for example, holding a book in my hand, and trying to extend my consciousness into the book - feel my hand tracing patterns on my cover, feel my fingers running over my pages, and so on.

Nothing came of this in itself, but I picked up a couple of skills - "knacks", "reflexes", or "habits" might be better words - that would become important later on.

That was the third beginning.  The second beginning was when I learned to trust my intuitions while practicing for the SAT, as recorded above.  The first beginning was when, at the age of nine, I happened across where my parents had stored volume 15 of my Childcraft set - "Guide for Parents".  So, surreptitiously, I read all the interesting parts.  The volume contained the great secret of the adult conspiracy; I do not recall being impressed.  I also read the description of what "teens" were like, how they were confused, and beset by anxieties about their social standing and their self-image, and how some of them were "rebellious" and "sullen".  "Yuck," said I.  I saw that these were basically obvious mistakes, and I decided I wasn't going to make them.  And it worked.  To this day, my primary source of information about the teenage experience is Buffy the Vampire Slayer.

Of course, falling out of school and having a broken will - missing out on the experiences, the acculturation, and even the emotions that ordinarily produce the late-twentieth-century cultural artifact called a "teenager" - may also have something to do with that.  But then again, maybe not.  "Refusing to behave like a stereotype" amounts to breaking a pattern, or not doing something I know is a mistake, a determination that hooks directly into my Specialist emotion.  Of course that consideration won out over everything else in the mental landscape.  What would you expect?

4.2: Unrationalization

Very long ago, in summer camp of either kindergarten or first grade, I remember refusing to play a game.  In this game, boys would line up, stand with their legs apart, and the boy at the end of the line would crawl through the line.  As the crawler passed, each boy would give him a spank.  Then the crawler wound up at the front of the line, the second-to-last boy was now last, and the game repeated.  And, believe it or not, this game from "The Marquis de Sade's Monster Fun Book" was sanctioned by the camp counselors (who were, big surprise, older boys).  When I refused to play, they told me to sit in a corner.

Later in life, up until the age of fourteen or so, I cherished this memory proudly.  It was a banner symbolizing my refusal to play negative-sum games; my refusal to take joy in someone else's pain, or play the sick social game where the reward of being hurt is the chance to hurt others.  It was a part of my self-image, my identity.

And then, one day, I realized the story wasn't historically true.  I had, in fact, remembered my motive along with the event, and it was nothing remotely like "Don't play negative-sum games."  I was too young to have that kind of moral philosophy; I had refused to participate because I didn't want to get hurt myself.  And then, I caught the feeling of catching myself in a rationalization.  I caught the feeling of having a piece of knowledge in the corner of my mind, but refusing to look at it squarely.

One of the forces that upholds "not thinking about it" is an unspoken anxiety about what happens if you do think about it.  Well, I'd thought about it, and it hadn't hurt me, and that gave me the courage to do it again.  And again.  And again, until it became a knack, and then a reflex, by which time I was proceeding to clean house generally.

Once I got the hang of it, this method probably accounted for at least fifty percent of my effective intelligence.  The occasional brilliant insight, the sudden leap from A to Z, is an ability I was born with; but the ability to go directly from A to B to C to D without a twenty-year timeout to rationalize some plausible-sounding mistake, the ability to get it right on a day-to-day basis, is the result of the mental reflexes that started with that knack.

4.3: Evolutionary psychology

One day, I ran across an article (35) about a man called "Robert Wright", a science called "evolutionary psychology", and a book called "Man:  The Moral Animal".  The particular article was about the evolutionary psychology of adultery; it also happened to mention the idea that the differences between male and female sex drives could be accounted for by the fact that, to produce a baby, a male only invests a night, while a female invests nine months.  And therefore the male evolutionary strategy is to sleep with everything that moves, while a female's evolutionary strategy is to be extremely picky.  This is oversimplifying down to the talk-show level - you'll note I say "male" and "female" rather than "man" and "woman" - but it remains the underlying truth, which the complexities only modify.

I latched onto evolutionary psychology like a starving suction cup; it was so obvious, in retrospect, that this was the source of emotion.  I was still right in the middle of my aforesaid mental housecleaning, so the scope of that housecleaning expanded suddenly and dramatically.

I obtained a copy of "The Moral Animal" in short order, and proceeded to learn many interesting things - although, if truth be told, I could probably have done without the book itself.  Once I had the basic concept, evolutionary psychology was just so obvious.  I understood - independently (36) - where the "male teenager" stereotype came from: stereotypical male teenagers are physiological adults, with the adult set of emotional instincts, in a social context that these instincts interpret as low social status - the bottom rungs of the hunter-gatherer tribe, with no opportunity for advancement within the system.  Well, in that situation, the instinctive response is to attempt to overthrow the current chief of the tribe - who, for obvious reasons, often winds up being identified with Dad.  (Meanwhile, of course, the parents are still operating on parental instincts, and major-league cross-wiring occurs.)  Or so I imagine, anyway, since I've never been through it personally.

That explanation is hardly The Adapted Mind, but then it doesn't need to be.  I wasn't trying to prove a scientific point; I was just trying to renovate the living daylights out of my emotional landscape and continue to ensure that I never, ever wound up as a teenager.  Even back then, I'd written more complex analyses, but who needs them?  The basic knowledge was more than enough to drive the final nails into the coffin.

And so, armed with an understanding of my own mind, I set out to eliminate the ugly parts.

4.4: Cleaning up the mental landscape

In the beginning, of course, I targeted the "special circumstances" emotions - the ones with blatantly adaptive responses to unusual external conditions.  (That wasn't a conscious decision.  Those were just the only emotions I noticed, the ones that were salient enough to target.)  I set out to eliminate rebelliousness, feelings of superiority, and above all, self-righteousness.  I identified self-righteousness, the smug feeling of moral justification, as evolution's "signature style" on the moral emotions; whenever I observed it, I set out to find the thoughts that had triggered the emotion and turn them off.

Does that sound dangerous?  Perhaps it is.  But from the beginning, I was aware of the Cultural Wisdom that says suppressing or repressing emotions is a bad idea.  (I have my doubts as to whether this is true, but on the other hand, I have my doubts as to whether this is false.  Never mind.)  In my case, the dominant concern was that both suppression and repression use up mental energy.  Neither would be sustainable for very long.  My goal, instead, was to "extend my consciousness" into my emotions - to control, or rather alter my emotions; not by expending willpower, but in the same way I flex a finger.  (37).  Then, as now, I focused not on self-control, but on self-alteration.

Does that sound dangerous?  That far back, it wasn't - not really.  I could safely target the moral emotions for wholesale elimination because they were special-circumstances emotions.  Some years later, I began to notice the existence of more generalized social emotions - the urge to convince other people that you're right, for example.  I didn't dare set out to eliminate that.  By that time, I'd formulated Algernon's Law, meaning that I was aware of the links between emotions and cognitive abilities, and switching off that social emotion might have reduced my intelligence with respect to persuading people.  Also, the social emotions were too ubiquitous.  They weren't just activated on special occasions, they formed part of the pattern binding ordinary life together.  When I finally did notice the existence of social emotions, I tried to notice their activations, and prevent them from doing anything noticeably stupid, but I didn't set out to eliminate them.

On the other hand, I did target the generalized habit of emotional investment in a belief.  Emotional investment results in an inability to emotionally accept other possibilities - the mind flinches away - and that is not something a futurist can afford.  (38).  Also, getting invested in an idea is a nasty habit on a purely cognitive level.

Initially, for example, I didn't tamper with my emotional investment in the Singularity because I was worried it would affect my energy levels.  As a result, for a couple of years, I failed to consciously admit the possibility of a nanowar.  The thought "Well, you could prevent a Singularity by wiping out all intelligent life in the Solar System" didn't form, because I had permitted myself to believe in the Singularity.  I also failed to concretely visualize the effects of public opposition, even though I knew, in the abstract, that it would exist; my mind flinched away, again because it led to outcomes that I wasn't willing to accept.  Well, I learned my lesson on that one.  Afterwards, I did my best to keep cognition about preferred outcomes completely clear of cognition about probable outcomes, with no consciously granted exceptions.

4.5: Messing with the deep stuff

Up until this point, I had remained pretty much normal.  The content of my mind was probably off the chart in a few directions - cleanliness, for one - but the rules remained the same.  If you'd suggested that I was not just unusual but eerie, or any other term that carried with it respect-for-weirdness, I would have instantaneously dismissed the suggestion as an emotionally appealing falsehood which derived any perceived plausibility from the smugness generated by the mental image of self-as-superior-to-others.  And I would have been essentially right.  (Of course, I wouldn't have used that many words unless I was actually talking to someone; the reflex itself runs directly from perception of bad-tasting emotional appeal to yuck.  (39).)

Up until eighteen or nineteen, my personality worked in basically the same way as everyone else's.  I remained normal through elementary school, through taking the SAT, through falling out of school, through discovering evolutionary psychology, through dedicating myself to the Singularity, through figuring out Algernon's Law, and through my breaking point.  And if you asked me, I'd have said so.

After all that, it was writing Coding a Transhuman AI 1.0 that did me in.

It was after writing the first version of CaTAI that my image of my mind as a single object began to break down.  Very, very slowly, I began to see some of the components, and the glue holding them together.  In theory, of course, I'd known perfectly well that the mind was built of modules.  After writing CaTAI, my actual perception of self started to break down.

The way I initially conceptualized it was "thinking without using the word 'I'", or "disintegration of the self-symbol".  Instead of attributing causality to a monolithic object, saying "I remembered the sky is blue", one might say "short-term memory says the sky is blue", or "short-term memory reconstructs an image of a blue sky in the visual cortex".  Two years after thinking up this discipline, I still can't actually do it for more than one slow mental sentence at a time; it's just too tedious.  I hope it gives you some of the flavor, but the important part isn't attributing pieces of thought to individual modules in a description written after the fact; the important part is being able to watch the mental event happen, piece by piece, affect by affect, flavor by flavor.

What I do amounts to having a painter's understanding of the mind.  A painter has the same visual neurology as any other human; ve just pays more conscious attention to the pieces of the visual field, and not just the holistic outputs.  That is, a painter can look at a physically 90-degree 3-D angle, and bypass the visual interpretation of "right angle" to observe the photographically 160-degree 2-D angle that gets put down on paper.  A painter pays attention to facts about lighting and shading that most of us never consciously notice.  Even though we have the same visual perceptions as the painter, we don't have names for what we see.  My self-awareness doesn't have any viewscreens or steering wheels that aren't part of the standard set - I think - but I have names for some of the tones and intuitions that are ordinarily assembled into monolithic "emotions" and "ideas".  That, and years of practice, have enabled me to form perceptual reflexes that let me see some tones and intuitions as conscious, discrete events.

Not to give the wrong impression, I should note that I am far from being able to do this all of the time.  We can move our fingers because we have a sensorimotor cortex devoted to placing finger motions under voluntary control.  If the mind doesn't have a built-in steering wheel for something, than you can't control it directly.  If the mind does have enough of a built-in viewscreen that you can watch the causal sequences, then sometimes you can single out an element of the sequence that you can control and use that to affect the end results.  All of this is much harder than it sounds; what I have, essentially, are one or two mental tricks that I can use on the two or three types of sequence I understand to achieve extremely blunt results with little or nothing in the way of quantitative or structural control.  And while I'm still working on it, I don't ever expect to achieve the kind of total control that the general theory implies.  I don't even expect to come close.

Those self-alteration tricks, however, are powerful enough that I'm reluctant to talk about them.  They're dangerous.  Not deadly dangerous to anyone with an ounce of common sense, and certainly not capable of causing hardware damage, but probably capable of causing major damage to a personality.  One of the generalized knacks I picked up lets me switch off arbitrary minor emotions.  There's a trick to disrupting the sequitur between a thought and the emotion that follows it.  It's a very simple and general trick, and once it's been used often enough to become a "knack", trying not to use it can be like trying not to think of the word "pickle".  Now, I can actually take a fairly good shot at not thinking of the word "pickle".  I don't know if my readers can.  So I'm not going to discuss that trick.

Likewise, I have no idea whether the same trick will work for intense emotions.  It seems unlikely, since I beguess (40) that intense emotions involve large, multiple-sequence causal interactions, and some of those sequences aren't ones which I can watch in slow motion.  But that's a moot point.  I don't know if the trick works on major emotions because I've never tried.  Positive or negative, intense emotions are rare enough, in my life, that I would never deliberately diminish one.  And to be blunt, I'm afraid to try.

4.6: What's it like to be you?

I am sufficiently intelligent to have completely avoided most or all of the pitfalls of youth, and I've cleaned enough dirt out of my mind that the thought of living in a completely open telepathic society doesn't disturb me.  And yet I still fall short of moral perfection, because I have far less mental energy than an ordinary human.

The human mind is scarily, unintuitively modular, as even a casual student of brain-damage cases knows.  Consider, for example, the phenomenon of anosognosia, in which a patient's left arm or leg is paralyzed and the patient doesn't know it.  The patient will deny any paralysis, rationalize vis "decision" not to move the arm, become angry if the confrontation continues, and so on.  This only happens with the left side of the body, never the right!

It's horrifying to contemplate the thought that your self-awareness, your rationality, is subject to that sort of selective damage.  It makes you wonder whether there are other blatantly obvious facts that we're all incapable of perceiving.

I am a neurohack, every bit as alien as any denizen of The Modular Brain or The Man Who Mistook His Wife for a Hat.  But that alienness expresses itself in an unintuitive, modular, dramatically inconsistent way.  I have a painter's self-awareness, a clean mind, and tricks of self-alteration; things which, if they were transferred to someone else, would probably result in nearly total "self-control"; I have beliefs which, all else being equal, would create an intense, energetic fanatic.  It doesn't work that way, for me, because I have a hardware deficit creating a lack of mental energy.  Nothing trumps hardware.  Period.  I can be a fanatic Singularitarian, reorganize my mind, and it won't ever give me as much mental energy as an ordinary human.  End of story.

Our folk self-image may contain the built-in delusion that the options and behaviors called willpower are the result of belief and mental discipline, as well as hardware mental energy levels - just as our folk self-image contains the illusion that the ability to know whether your left arm is paralyzed is the result of general rationality and reasoning ability.  In my case, the illusion is broken.  There are a lot of characteristics, behavior patterns, et cetera, which - from my front-row seat - would appear to be calculated as the direct result of hardware mental energy levels.  Everything else has little or no effect.

Given a task, I still have an enormous amount of trouble actually sitting down and doing it.  (Yes, I'm sure it's unpleasant for you too.  Bump it up by an order of magnitude, maybe two, then live with it for eight years.)  My energy deficit is the result of a false negative-reinforcement signal, not actual damage to the hardware for willpower; I do have the neurological ability to overcome procrastination by expending mental energy.  I don't dare.  If you've read the history of my life, you know how badly I've been hurt by my parents asking me to push myself.  I'm afraid to push myself.  It's a lesson that has been etched into me with acid.

And yes, I'm good enough at self-alteration to rip out that part of my personality, disable the fear, but I don't dare do that either.  The fear exists for a reason.  It's the result of a great deal of extremely unpleasant experience.  Would you disable your fear of heights (41) so that you could walk off a cliff?  I can alter my behavior patterns by expending willpower - once.  Put a gun to my head, and tell me to do or die, and I can do.  Once.  Put a gun to my head, tell me to do or die, twice, and I die.  It doesn't matter whether my life is at stake.  In fact, as I know from personal experience, it doesn't matter whether my home planet and the life of my entire species is at stake.  If you can imagine a mind that knows it could save the world if it could just sit down and work, and if you can imagine how that makes me feel, then you have understood the single fact that looms largest in my day-to-day life.

In the abstract, I know perfectly well that it's impossible for a Countersphexist - that's my Specialty - to work 16-hour-days.  Heck, it's impossible for an ordinary human to work 16-hour-days.  But could I work one more hour a day without damaging myself?  Am I five minutes to the wrong side of the border between legitimate concern about burning myself out, and the phenomenon known as laziness?  It's like a surreal, miniaturized version of the standard human dilemma.  And, of course, procrastination is itself a process that dissipates mental energy.  I'm trying to alter this particular vicious cycle, but none of what I've described is the result of a conscious decision.  It's the result of sediment - reflex and cached decision - that has built up over years, plus some direct Algernic vulnerabilities.  Those are not easy to alter, and it's also not easy to untangle the bad habits from the learned safeguards.

On the plus side, despite all the problems, I do have my Singularitarianism and my knacks at self-alteration, and that's what makes life livable.  In the back of my mind, there is an eternal keen of something very like despair.  It doesn't drive me insane, first because I'm used to it, and second because I deliberately don't have the expectation that it will drive me insane.  I can avoid the tendency to dramatize, to attach moral valence to what is simply a neurological false signal.  The brain is modular, and when modules are perturbed, the results are nonintuitive.  Because I don't force coherence when none exists, I can function as a human with one module perturbed, instead of my mind being the devastated ruin that folk psychology would predict.

A personality is a huge thing.  It starts with the panhuman psychology, the underlying emotional and intuitive hardware that determines which cognitive structures are possible.  On top of that is the sediment - cognitive reflexes, emotional resonances - formed by experience, by mental habits that are the result of pure happenstance or even conscious decision, by unquestioned ideas we took with us out of childhood, by fantasies, by the scenarios we keep running through mentally, and so on.  This vast thought-cache is not evil; caching is what enables us to think a thought in a second instead of a year... but a shocking amount of personality-substance does turn out to be the result of habits formed at very early ages.

And a large amount of my personality-substance has emotional resonance or intuitive structure that hooks directly into my Algernic tone - especially all the intense parts.  The mountains on my mental landscape - the arguments that resonate so strongly that I accept them as final - are variations on that one Algernic theme.

5: What are your plans for the future?

Finish writing Coding a Transhuman AI 2, including the planned sections on "Cognition", "Design", and "Development".  When complete, I should know what the initial stage should be, how that initial stage fits into the larger scheme, and the specific design tasks required to implement the initial stage.  I should have a design document good enough to show to a project team.  And then the only thing standing between humanity and the Singularity will be money, time, and computing power.

I'm no longer sure that open-sourcing the code and starting an AI industry is necessary.  The more I understand AI, the more I feel that it's not about huge amounts of industry effort or huge amounts of computing power.  It's about understanding what you're doing.  Everything else is secondary.


My plan for the future is to complete my understanding of intelligence, finish the design of the initial development stages.  If the Singularity Institute doesn't have enough funding, I'll go looking for more funding; or else hack up a version of the AI that does something cool, both to attract sponsorship and to increase my understanding of AI.  Once there's enough funding, I'll pull a project team together and start coding.  We'll write the first-stage AI, then write a seed AI using that first-stage AI as glue.  We'll teach the seed AI how to think, how to understand code, how to redesign verself, and hopefully give ver a personal philosophy that involves being friendly to humanity.

And then one day ve makes a key breakthrough, the Singularity occurs, and I'm outta here.

5.1: What about your life?

You mean, what about me, Eliezer, the person, not Eliezer S. Yudkowsky, the Singularitarian?  What about the real me?

Eliezer S. Yudkowsky is the real me.  This is who I am.  I can understand the tendency to view one's hobbies, likes and dislikes, random quirks, friends and family, anything and everything outside "work" as being the real you.  And yet, I don't feel the need to separate my life into "personal" and "professional".  I know I'm going against a cherished cultural myth, but I just don't.  Logic is logic.  Mind is mind.  The rational choice is the rational choice.  For me, the things usually called personal are organized the same way as everything else, subsumed into the smooth, continuous, goal-and-subgoal logic of the Singularity.

Yes, it's necessary that I spend some time reading science fiction - but it really is necessary.  Even for a normal high-energy human it would be necessary.  It still serves the Singularity.  Perhaps I'm taking too much time off, crossing the border between necessity and rationalization.  (43).  But trying to function with no time off just wouldn't work.

According to perceived cultural wisdom, viewing your downtime as a subgoal of your uptime detracts from the efficiency of the downtime, and you're happier if you value your personal life for its own sake.  Unlike most twentieth-century emotional mythology, this one is partially true; there is, in fact, a direct sequitur from {cognitive: pleasure X is a subgoal} to {affective: reduce the enjoyment of X}.  (44).  But there are ways to get around that - although, admittedly, I'm still working on some of them.

In the final analysis, I'm not afraid to have less fun; it may decrease available willpower, but it doesn't expend willpower.  The prospect of purposeful enjoyment diminishing enjoyment isn't something I flinch away from, an argument that I accept as final.  The argument I do accept as final is that if I maintain my commitment to rationality, I may someday learn how solve the problem.

5.2: Do you really think you can change the world?

Well, yes.  After all, if I shouldn't believe that I can do it - or at least, that I have enough of a chance to make the effort worthwhile - who should?

That, at any rate, would be the argument I'd use if I were trying to set up a socially acceptable set of emotional bindings under which "I can change the world" sounds like a modest endeavor.  "You can assign a suitably humble estimate to my chances of success as an individual, but if nobody tried, we'd be out of luck, wouldn't we?"

That's not what I tell myself.

Algernon's Law says that any simple (45) major intelligence enhancement will turn out to be a net evolutionary disadvantage.  (46).  The flip side of Algernon's Law is that there are some levels of performance that can't be achieved without the evolutionary disadvantage.  If there were any way to duplicate the performance level of a Specialist without the downside, my SAT scores would be as common as mud; those performance levels would be part of the standard human neurology, like symbolic language or three-dimensional vision.  I therefore expect that in any of my areas of Specialization - causal analysis, combinatorial design, reflectivity - I will have significantly more raw processing power than anyone except another Specialist.

Just because ten thousand dedicated researchers hit a problem and bounce, it is not a licensable conclusion that a Specialist will bounce off as well.  A truly "impossible" problem often takes me a while to solve; I have to slog away at it steadily, whereas my usual rule is that I can see an answer instantly if I can see it at all.  Conversely, if my level of processing power is enough to make slow but steady progress on a problem, then I would expect any non-Specialist to hit the problem and bounce.  Which, to some extent, explains the travails suffered by AI.

The existence of Specialists  expands the range of problems that humanity can solve.  And because I don't know of any other Countersphexists operating in the fields of AI, transhumanism, or Singularitarianism (47), or of any other Countersphexists at all (48), then, for an important class of problems, I expand the range of problems that can be solved.  In particular, I am referring to the problem of AI.

Oh, don't get me wrong - I'm sure AI would be solved eventually.  In about 2020 CRNS (49), the weight of accumulated cognitive science and available computing power would disintegrate the ideological oversimplifications and create enough of a foothold in the problem to last humanity the rest of the way.  It would be absurd to claim to represent the difference between a solvable and unsolvable problem in the long run, but one genius can easily spell the difference between cracking the problem of intelligence in five years and cracking it in twenty-five - or to put it another way, the difference between a seed AI created immediately after the invention of nanotechnology, and a seed AI created by the last dozen survivors of the human race huddled in a survival station, or some military installation, after Earth's surface has been reduced to a churning mass of goo.

That's why I matter, and that's why I think my efforts could spell the difference between life and death for most of humanity, or even the difference between a Singularity and a lifeless, sterilized planet.  I don't mean to say, of course, that the entire causal load should be attributed to me; if I make it, then Ed Regis or Vernor Vinge, both of whom got me into this, would equally be able to say "My efforts made the difference between Singularity and destruction."  The same goes for Brian Atkins, and Eric Drexler, and so on.  History is a fragile thing.  So are our causal intuitions, where linear chains of dependencies are concerned.  Nonetheless, I think that I can save the world, not just because I'm the one who happens to be making the effort, but because I'm the only one who can make the effort.  And that is why I get up in the morning.