(At this point, I fear that I must recurse into a subsequence; but if all goes as planned, it really will be short.)

    I once lent Xiaoguang "Mike" Li my copy of "Probability Theory: The Logic of Science".  Mike Li read some of it, and then came back and said:

    "Wow... it's like Jaynes is a thousand-year-old vampire."

    Then Mike said, "No, wait, let me explain that—" and I said, "No, I know exactly what you mean."  It's a convention in fantasy literature that the older a vampire gets, the more powerful they become.

    I'd enjoyed math proofs before I encountered Jaynes.  But E.T. Jaynes was the first time I picked up a sense of formidability from mathematical arguments.  Maybe because Jaynes was lining up "paradoxes" that had been used to object to Bayesianism, and then blasting them to pieces with overwhelming firepower—power being used to overcome others.  Or maybe the sense of formidability came from Jaynes not treating his math as a game of aesthetics; Jaynes cared about probability theory, it was bound up with other considerations that mattered, to him and to me too.

    For whatever reason, the sense I get of Jaynes is one of terrifying swift perfection—something that would arrive at the correct answer by the shortest possible route, tearing all surrounding mistakes to shreds in the same motion.  Of course, when you write a book, you get a chance to show only your best side.  But still.

    It spoke well of Mike Li that he was able to sense the aura of formidability surrounding Jaynes.  It's a general rule, I've observed, that you can't discriminate between levels too far above your own. E.g., someone once earnestly told me that I was really bright, and "ought to go to college".  Maybe anything more than around one standard deviation above you starts to blur together, though that's just a cool-sounding wild guess.

    So, having heard Mike Li compare Jaynes to a thousand-year-old vampire, one question immediately popped into my mind:

    "Do you get the same sense off me?" I asked.

    Mike shook his head.  "Sorry," he said, sounding somewhat awkward, "it's just that Jaynes is..."

    "No, I know," I said.  I hadn't thought I'd reached Jaynes's level. I'd only been curious about how I came across to other people.

    I aspire to Jaynes's level.  I aspire to become as much the master of Artificial Intelligence / reflectivity, as Jaynes was master of Bayesian probability theory.  I can even plead that the art I'm trying to master is more difficult than Jaynes's, making a mockery of deference.  Even so, and embarrassingly, there is no art of which I am as much the master now, as Jaynes was of probability theory.

    This is not, necessarily, to place myself beneath Jaynes as a person—to say that Jaynes had a magical aura of destiny, and I don't.

    Rather I recognize in Jaynes a level of expertise, of sheer formidability, which I have not yet achieved.  I can argue forcefully in my chosen subject, but that is not the same as writing out the equations and saying:  DONE.

    For so long as I have not yet achieved that level, I must acknowledge the possibility that I can never achieve it, that my native talent is not sufficient.  When Marcello Herreshoff had known me for long enough, I asked him if he knew of anyone who struck him as substantially more natively intelligent than myself.  Marcello thought for a moment and said "John Conway—I met him at a summer math camp."  Darn, I thought, he thought of someone, and worse, it's some ultra-famous old guy I can't grab.  I inquired how Marcello had arrived at the judgment.  Marcello said, "He just struck me as having a tremendous amount of mental horsepower," and started to explain a math problem he'd had a chance to work on with Conway.

    Not what I wanted to hear.

    Perhaps, relative to Marcello's experience of Conway and his experience of me, I haven't had a chance to show off on any subject that I've mastered as thoroughly as Conway had mastered his many fields of mathematics.

    Or it might be that Conway's brain is specialized off in a different direction from mine, and that I could never approach Conway's level on math, yet Conway wouldn't do so well on AI research.


    ...or I'm strictly dumber than Conway, dominated by him along all dimensions.  Maybe, if I could find a young proto-Conway and tell them the basics, they would blaze right past me, solve the problems that have weighed on me for years, and zip off to places I can't follow.

    Is it damaging to my ego to confess that last possibility?  Yes.  It would be futile to deny that.

    Have I really accepted that awful possibility, or am I only pretending to myself to have accepted it?  Here I will say:  "No, I think I have accepted it."  Why do I dare give myself so much credit?  Because I've invested specific effort into that awful possibility.  I am blogging here for many reasons, but a major one is the vision of some younger mind reading these words and zipping off past me.  It might happen, it might not.

    Or sadder:  Maybe I just wasted too much time on setting up the resources to support me, instead of studying math full-time through my whole youth; or I wasted too much youth on non-mathy ideas.  And this choice, my past, is irrevocable.  I'll hit a brick wall at 40, and there won't be anything left but to pass on the resources to another mind with the potential I wasted, still young enough to learn.  So to save them time, I should leave a trail to my successes, and post warning signs on my mistakes.

    Such specific efforts predicated on an ego-damaging possibility—that's the only kind of humility that seems real enough for me to dare credit myself.  Or giving up my precious theories, when I realized that they didn't meet the standard Jaynes had shown me—that was hard, and it was real.  Modest demeanors are cheapHumble admissions of doubt are cheap.  I've known too many people who, presented with a counterargument, say "I am but a fallible mortal, of course I could be wrong" and then go on to do exactly what they planned to do previously.

    You'll note that I don't try to modestly say anything like, "Well, I may not be as brilliant as Jaynes or Conway, but that doesn't mean I can't do important things in my chosen field."

    Because I do know... that's not how it works.

    New Comment
    357 comments, sorted by Click to highlight new comments since:
    Some comments are truncated due to high volume. (⌘F to expand all)Change truncation settings

    In a few years, you will be as embarrassed of these posts as you are today of your former claims of being an Algernon, or that a logical paradox would make an AI go gaga, the tMoL argumentation you mentioned the last days, the Workarounds for the Laws of Physics, Love and Life Just Before the Singularity and so on and so forth. Ask yourself: Will I have to delete this, too ?

    And the person who told you to go to college was probably well-meaning, and not too far from the truth. Was it Ben Goertzel ?

    Despite all fallibility of memory, I would be shocked to learn that I had ever claimed that a logical paradox would make an AI go gaga. Where are you getting this from?

    Ben's never said anything like that to me. The comment about going to college was from an earnest ordinary person, not acquainted with me. And no, I didn't snap at them, or laugh out loud; it was well-intentioned advice. Going to college is a big choice for a lot of people, and this was someone who met me, and saw that I was smart, and thought that I seemed to have the potential to go to college.

    Which is to imply that if there's a level above Jaynes, it may be that I won't understand it until I reach Jaynes's level - to me it will all just look like "going to college". If I recall my timeline correctly, I didn't comprehend Jaynes's level until I had achieved the level of thinking naturalistically; before that time, to achieve a reductionist view of intelligence was my whole aspiration.

    Although I've never communicated with you in any form, and hence don't know what it's like for you to answer a question of mine, or correct a misconception (you have, but gradually), or outright refute a strongly held belief...or dissolve a Wrong Question...

    ...You're still definitely the person who strikes me as inhumanly genius - above all else.

    Unfortunately for my peace of mind and ego, people who say to me "You're the brightest person I know" are noticeably more common than people who say to me "You're the brightest person I know, and I know John Conway". Maybe someday I'll hit that level. Maybe not.

    Until then... I do thank you, because when people tell me that sort of thing, it gives me the courage to keep going and keep trying to reach that higher level.

    Seriously, that's how it feels.

    I think maybe Being the Smartest Person is a fundamentally bad, unhelpful motivator, and you should get some cognitive therapy. Of course, you would immediately conclude (correctly) that you are smarter than your mental health professional and stop listening (stupidly and non-volitionally) to them. So this is probably a road you're going to have to walk. Here's hoping you don't have a horrible self- or other-destructive flameout.

    You are the brightest person I know. And I know Dan Dennett, Max Tegmark, Robert Trivers, Marcello, Minsky, Pinker and Omohundro.

    Unfortunately, those are non-math geniuses, so that speaks for only some sub-areas of cognition which, less strictly categorizable than the clearly scalable domain of math, are not subject to your proposed rule of "one standard deviation above you they blurr"

    "Know" in the sense EY used it != have read, watched interviews, etc. I took it to mean more personal interaction (even if through comments online).
    Especially since "know of" exists as a common phrase to cover the meaning "have read, watched interviews, etc."

    I have had classes with them, asked questions. and met them personally. I should have anticipated disbelief. And yes, I didn't notice that I categorized Marcello as non-math, sorry Marcello!

    Oh. Cool! Less disbelief, more illusion of transparency.

    If a randomly selected person says, "I know X (academically) famous people." I myself usually assume through impersonal means.

    Update'd. Carry on :D

    Non-math geniuses who grok and advocate for unpopular reductionism are in one sense greater than mere superheroes who know the math.
    In another sense, non-math geniuses advocating for reductionism are no better than the anti-vaccine lobby.
    What sense is that?
    The sense in which they did not come about their beliefs based on starting with sane priors which did not presuppose reductionism, and then update on evidence until they independently discovered reductionism. I disagree with the grandparent, however: I believe that (most) non-math-geniuses advocating for reductionism are more akin to Einstein believing in General Relativity before any novel predictions had been verified: recognizing the absurdity of all other proposed hypotheses is another way of coming about the correct beliefs.
    The "absurdity" of non-reductionism seems to have evaded Robert Laughlin, Jaorn Lanier and a bunch of other smart people.
    I did not say that non-reductionism is absurd. I said that "recognizing the absurdity of all other proposed hypotheses is another way of coming about the correct beliefs". Nonetheless, I do think that non-reductionism is absurd. I cannot imagine a universe which is not reductionistic. Can you explain to me how it might work? Edit: I googled "Robert Laughlin Reductionism" and actually found a longish paper he wrote about reductionism and his beliefs. I have some criticisms: Yudkowsky has a great refutation of using the description "emergent", at The Futility of Emergence, to describe phenomenon. From there: Further down in the paper, we have this: Every time he makes the specific claim that reductionism makes worse predictions than a belief in "emergent phenomenon" in which "organizational structure" is an additional property that all of reality must have, in addition to "mass" and "velocity", he cites himself for this. He also does not provide any evidence for non-reductionism over reductionism; that is, he cannot name a single prediction where non-reductionism was right, and reductionism was wrong. He goes on to say that reductionism is popular because you can always examine a system by looking at its internal mechanisms, but you can't always examine a system by looking at it from from a "higher" perspective. A good example, he says, is genetic code: to assume that dna is actually a complete algorithmic description of how to build a human body is an illogical conclusion, according to him. He would rather suppose that the universe contains rules like "When a wavefunction contains these particular factorizations which happen not to cancel out, in a certain organizational structure, use a different mechanism to decide possible outcomes instead of the normal mechanism" than suppose that the laws of physics are consistent throughout and contain no such special cases. From the standpoint of simplicity, reductionism is simpler than non-reductionism, since non-redu
    It's a bit of an aside to your main point, but there are good arguments to support the assertion that DNA is only a partial recipe for an organism, such as a human. The remaining information is present in the environment of the mothers' womb in other forms - for example, where there's an ambiguity in the DNA with regards to the folding of a certain protein, other proteins present in the womb may correct any incorrectly folded samples. To look at your main point; if I were to present an argument against reductionism, I would point to the personal computer. This is a device constructed in order to run software; that is, to follow a list of instructions that manipulate binary data. Once you have a list of all the instructions that the computer can follow, and what these instructions do, a thorough electrical analysis of the computer's circuitry isn't going to provide much new information; and it will be a lot more complicated, and harder to understand. There's a conceptual point, there, at the level of individual software instructions, where further reductionism doesn't help to understand the phenomenon, and does make the analysis more complicated, and harder to work with. A thorough electrical analysis is, of course, useful if one wishes to confirm that the stated behaviour of the basic software commands is both correctly stated, and free of unexpected side-effects. However, an attempt to describe (say) the rendering of a JPEG image in terms of which transistors are activated at which point is likely a futile exercise.
    Well, yes - but that arises from the fact that such devices are man-made, and (out of respect to our brains' limitations) designed to isolate the layers of explanation from one another - to obviate the need for a fully reductionistic account. The argument will not apply to things not man-made.
    The entire science of psychology is based on the idea that it is useful to apply high-level rules to the neural functioning of the human brain. If I decide to eat a cookie, then I explain it in high-level terms; I was hungry, the cookie smelt delicious. An analysis in terms of the effect of airborne particles originating from the cookie on my nasal passages, and subsequent alterations in the pattern of neural activations in my brain, can give a far more complicated answer to the question of why I ate the cookie; but, again, I don't see how such a more complicated analysis would be better. If I want to understand my motivations more fully, I can do so in terms of mental biases, subconscious desires, and so forth; rather than a neuron-by-neuron analysis of my own brain. And while it is technically true that I, as a human, am man-made (specifically, that I was made by my parents), a similar argument could be raised for any animal. Such situations are rare, but not entirely unknown.
    I disagree with your entire premise. I think we should pin down this concept of "levels of perspective" with some good jargon at some point, but regardless... You can look at a computer from the level of perspective of "there are windows on the screen and I can move the mouse around. I can manipulate files on the hard drive with the mouse and the keyboard, and those changes will be reflected inside information boxes in the windows." This is the perspective most people see a computer from, but it is not a complete description of a computer (i.e. if someone unfamiliar with the concept of computers heard this description, they could not build a computer from base materials.) You might also see the perspective, "There are many tiny dots of light on a flat surface, lit up in various patterns. Those patterns are caused by electricity moving in certain ways through silica wires arranged in certain ways." This is, I think, one level lower, but an unfamiliar person could not build a computer from scratch from this description. Another level down, the description might be: "There is a CPU, which is composed of hundreds of thousands of transistors, arranged into logic gates such that when electricity is sent through them you can perform meaningful calculations. These calculations are written in files using a specific instruction set ("assembly language"). The files are stored on a disk in binary, with the disk containing many cesium atoms arranged in a certain order, which have either an extra electron or do not, representing 1 and 0 respectively. When the CPU needs to temporarily store a value useful in its calculations, it does so in the RAM, which is like the disk except much faster and smaller. Some of the calculations are used to turn certain square-shaped lights on a large flat surface blink in certain ways, which provides arbitrary information to the user". We are getting to the point where an unfamiliar human might be able to recreate a computer from scratch, and th
    That's a fusion of reductionism and determinism. Reductionism ins't necessarily false in an indeterministic universe. What is more pertinent is being able to predict higher level properties and laws from lower level properties and laws. (synchronously, in the latter case).
    No it isn't? I did not mean you would be able to make predictions which came true 100% of the time. I meant that your subjective anticipation of possible outcomes would be equal to the probability of those outcomes, maximizing both precision and accuracy.
    Yes it is. "A property of a system is said to be emergent if it is in some sense more than the "sum" of the properties of the system's parts. An emergent property is said to be dependent on some more basic properties (and their relationships and configuration), so that it can have no separate existence. However, a degree of independence is also asserted of emergent properties, so that they are not identical to, or reducible to, or predictable from, or deducible from their bases. The different ways in which the independence requirement can be satisfied lead to various sub-varieties of emergence." -- WP Still deterinism, not reductionism. In a universe where *1aTthere are lower-level-properties .. *1b operating according to a set of deterministic laws. *2a There are also higher-level properties.. *2b irreducible to and unpredictable from the lower level properties and laws... *2c which follow their own deterministic laws. You would be able to predict the future with complete accuracy, given both sets of laws and two sets of starting conditions. Yet the universe being described is explicitly non-reductionistic.
    I'm a bit confused. What exactly defines a "higher-level" property, if not that it can be reduced to lower-level properties?
    eg: being macrscopic, featuring only in the special sciences
    This all this means is that, in addition to the laws which govern low-level interactions, there are different laws which govern high-level interactions. But they are still laws of physics, they just sound like "when these certain particles are arranged in this particular manner, make them behave like this, instead of how the low-level properties say they should behave". Such laws are still fundamental laws, on the lowest level of the universe. They are still a part of the code for reality. But you are right: Which is what I said: Ergo, a reductionistic universe is also deterministic from a probabilistic standpoint, i.e. the lowest level properties and laws can tell you exactly what to anticipate, and with how much subjective probability.
    Microphysical laws map microphysical states to other microphysical states.Top-down causation maps macrophysical states to microphysical states. In the sense that they are irreducible, yes. In the sense that they are concerned only with microphyics, no. "Deterministic" typically means that an unbounded agent will achieve probabilities of 1.0.
    Can you name any examples of such a phenomenon? Oh, well in that case quantum physics throws determinism out the window for sure. I still think there's something to be said for correctly assigning subjective probabilities to your anticipations such that 100% of the time you think something will happen with a 50% chance, it happens half the time, i.e. you are correctly calibrated. An unbounded agent in our universe would be able to achieve such absolutely correct calibration; that's all I meant to imply.
    You are right; my example was a bad one, and it does not support the point that I thought it supported. The mere fact that something takes unreasonably long to calculate does not mean that it is not an informative endeavour. (I may have been working from a bad definition of reductionism). Um. I suspect that this may have been poorly phrased. If I have a lump of carbon, quite a bit of water, and a number of other elements, and I just throw them together in a pile, they're unlikely to do much - there may be a bit of fizzing, some parts might dissolve in the water, but that's about it. Yet if I reorganise the same matter into a human, I have an organisation of matter that is able to enter into a debate about reductionism; which I don't think can be predicted by looking at the individual chemical elements alone. But that behaviour might still be predictable from looking at the matter, organised in that way, at its most basic level of perspective (given sufficient computing resources). Hence, I suspect that it is not a counter-example.
    Not true. There is a reason no one uses quarks to describe chemistry. Its futile to describe whats happening in a superfluid helium in terms of individual particle movement. Far better to use a two fluid model, and vortices.
    Let me amend that: the argument will not necessarily apply to things not man-made. There is a categorical difference in this respect between man-made things and the rest, and my intent was to say: "if you're going to put up an argument against reductionism, don't use examples of man-made things". Whereas we have good reasons to bar "leaky abstractions" from our designs, Nature labors under no such constraint. If it turns out that some particular process that happens in a superfluid helium can be understood only by referring to the quark level, we are not allowed to frown at Nature and say "oh, poor design; go home, you're drunk". For instance, it turns out we can almost describe the universe in the Newtonian model with its relatively simple equations, a nice abstraction if it were non-leaky, but anomalies like the precession of Mercury turn up that require us to use General Relativity instead, and take it into account when building our GPS systems. The word "futile" in this context strikes me as wishful thinking, projecting onto reality our parochial notion of how complicated a reductionistic account of the universe "should" be. Past experience tells us that small anomalies sometimes require the overthrow of entires swathes of science, in the name of reductionism: there keep turning up cases where science considers it necessary, not futile, to work things out in terms of the lower levels of description.
    I think you are making a bad generalization when you turn to Newtonian mechanics vs. general relativity. There are important ways in which mesons and hadron are emergent from quarks that have no correspondence to the relationship between Newtonian mechanics and GR. As length scales increase, quarks go from being loosely bound fundamental degrees of freedom to not-even-good-degrees-of-freedom. At 'normal' length scales, free quarks aren't even allowed. The modern study of materials is also full of examples of emergence (it underlies much work on renormalization groups), although its farther from my expertise so the only example to spring to mind was liquid helium.
    As an aside to an aside, I wonder how much information about the DNA reading frame could in principle be extracted from the DNA of a female organism, given the knowledge (or the assumption) that mature females can gestate a zygote? Almost all possible reading frames would be discardable on the grounds that the resulting organism would not be able to gestate a zygote, of course, but I don't have any intuitive sense of how big the remaining search space would be. And as a nod towards staying on topic: Well, it will, and it won't. If what I mostly care about is the computer's behavior at the level of instructions, then sure, understanding the instructions gets me most of the information that I care about. Agreed. OTOH, if what I mostly care about is the computer's behavior at the level of electrical flows through circuits (for example, if I'm trying to figure out how to hack the computer without an input device by means of electrical induction, or confirm that it won't catch fire in ordinary use), then a thorough electrical analysis of the computer's circuitry provides me with tons of indispensible new information. What counts as "information" in a colloquial sense depends a lot on my goals. It might be useful to taboo the word in this discussion.
    My intuition says "very, very big". Consider: depending on womb conditions, the percentage of information expressed in the baby which is encoded in the DNA might change. As an extreme example, consider a female creature whose womb completely ignores the DNA of the zygote, creating instead a perfect clone of the mother. Such an example makes it clear that the search space is at least as large as the number of possible female creatures that are able to produce a perfect clone of themselves. I accept your point. Such an analysis does provide a more complete view of the computer, which is useful in some circumstances.
    Sure, I agree that one permissible solution is a decoder which produces an organism capable of cloning itself. And while I'm willing to discard as violating the spirit of the thought experiment decoder designs which discard the human DNA in its entirety and create a predefined organism (in much the same sense that I would discard any text-translation algorithm that discarded the input text and printed out the Declaration of Independence as a legitimate translator of the input text), there's a large space of possibilities here.
    Would you be willing to consider, i.e. not discard, a decoder that used the human DNA as merely a list of indexes, downloading the required genes from some sort of internal lookup table? By changing the lookup table, one can dramatically change the resulting organism; and having a different result for every viable human DNA is merely a resut of having a large enough lookup table. It would be, to extend your metaphor, like a text-translation algorithm that returned the Declaration of Independance if given as input Alice in Wonderland, and returned Alice in Wonderland if given Hamlet.
    (considers) I would like to say "no", but can't think of any coherent reason to discard such a design. Yeah, OK; point made.
    I did not say that non-reductionism is absurd. I said that "recognizing the absurdity of all other proposed hypotheses is another way of coming about the correct beliefs". Nonetheless, I do think that non-reductionism is absurd. I cannot imagine a universe which is not reductionistic. One formulation of reductionism is that natural laws can be ordered in a hierarchy, with the higher-level laws being predictable from, or reducible to, the lower ones. So emergentism, in the cognate sense, not working would be that stack of laws failing to collapse down to the lowest level. There's two claims there: one contentious, one not. That there are multiply-realisable, substrate-independent higher-level laws is not contentious. For instance, wave equations have the same form for water waves, sound waves and so on. The contentious claim is that this is ipso facto top-down causation. Substrate-independent laws are still reducible to substrates, because they are predictable from the behaviour of their substrates. I don't see how that refutes the above at all. For one thing, Laughlin and Ellis do have detailed examples of emergent laws (in their rather weak sense of "emergent"). For another, they are not calling on emergence itself as doing any explaining. "Emergence isn't explanatory" doesn't refute "emergence is true". For a third, I don't see any absurdity here. I see a one-word-must-have-one-meaning assumption that is clouding the issue. But where a problem is so fuzzilly defined that it is hard even to identify the "sides", then one can't say that one side is "absurd". Neither are supposed to make predictions. Each can be considered a methodology for finding laws, and it is the laws that do the predicting. Each can also be seen as a meta-level summary fo the laws so far found. EY can't do that for MWI either. Maybe it isn't all about prediction. That's robustly true. Genetic code has to be interpreted by a cellular environment. There are no self-decoding codes. Reudcti
    Marcello is non-math?

    For what it's worth, I've worked on a project and had lunch with Conway, and your ideas seem more prescient than his. But being a mathematician, I know people who are in turn far above Conway's level.

    So how does it work, in your opinion? Because “I may not be as brilliant as Jaynes or Conway, but that doesn't mean I can't do important things in my chosen field,” sounds suspiciously similar to how Hamming asserts that it works in “You and Your Research.” I guess you have a different belief about how doing important things in your chosen field works, but I don't see that you've explained that belief here or anywhere else that I've seen.

    I don't suppose Marcello is related to Nadja and Josh Herreshoff?

    I don't know if it helps, but while I've appreciated the things I've learned from you, my limited interaction with you hasn't made me think you're the brightest person I know. I think of you as more or less at my level — maybe a couple of standard deviations above or below, I can’t really tell. Certainly you're sharp enough that I'd enjoy hanging out with you. (Let me know the next time you're in Argentina.)

    P.S. the impugnment of your notability has now been removed from your Wikipedia page, apparently as a result of people citing you in their papers.

    I too would like to hear "how it works," because if I don't know how Eliezer thinks it works, it just sounds like he's defining the problem of Being a Great Researcher in the most intimidating way possible. Whatever way that may be. Inflating the problem like that is bad practice, for much the same reason that cheap gestures of humility are bad practice. I'm commenting on a two-year-old post, so I guess I shouldn't expect a response, but this post is linked from the getting-started page, so I was a bit disappointed that it ended with what looks a lot like a handwave at humility.

    Wait wait wait wait. Eliezer...are you saying that you DON'T know everything????

    ~runs off and weeps in a corner in a fetal position~

    CatAI (1998): "Precautions"/"The Prime Directive of AI"/"Inconsistency problem".

    My memory may fail me, and the relevant archives don't go back that far, but I recall Ben (and/or possibly other people) suggesting you going to college, or at least enroll for a grad program in AI, on the Extropy chat list around 1999/2000. I think these suggestions were related to, but not solely based on, your financial situation at that time (which ultimately led to the creation of the SIAI, so maybe we should be glad it turned out the way it did, even if, in my opinion, following the advice would have been beneficial to you and your work.)


    I definitely see the "levels" phenomenon very often. Most people I meet who see me play a musical instrument (or 5 or 10 different ones) think I must be a genius at music - unless they're a musician, then they recognize me as an amateur with enough money to buy interesting instruments and enough skill to get a basic proficiency at them quickly.

    And even with standard measures of intellect like rationality or math... I don't know that many of my friends who have read any of this blog would recognize you as being smarter than me, despite the fact that you're enough levels above me that my opinion of you is pretty much what "Not You" said above.

    I can keep up with most of your posts, but to be able to keep up with a good teacher, and to be that good teacher, is a gap of at least a few levels. But aspiring to your level (though I may not reach it) has probably been the biggest motivator for me to practice the art. I certainly won't be the one who zips by you, but you've at least pulled me up to a level where I might be able to guide one who will down a useful path.

    Up to now there never seemed to be a reason to say this, but now that there is:

    Eliezer Yudkowsky, afaict you're the most intelligent person I know. I don't know John Conway.

    Your faith in math is misplaced. The sort of math smarts you are obsessed with just isn't that correlated with intellectual accomplishment. For accomplishment outside of math, you must sacrifice time that could be spent honing your math skills, to actually think about other things. You could be nearly the smartest math type guy anyone you meet know, and still not accomplish if math is not the key to your chosen subject.

    It's interesting, actually. You're motivated by other peoples' low opinions of you -- this pressure you feel in your gut to prove Caledonian et al wrong -- so you've taken that is probably fairly standard human machinery and tried to do something remarkable with it.

    My question is, are you still motivated by the doubt you feel about your native abilities, or have you passed into being compelled purely by your work?

    Perhaps the truly refulgent (before they had so become) reached a progression tipping point at which they realized (right or wrong, ironically) that they were essentially beyond comparison, and hence stopped comparing.

    Then they could allocate the scarce resources of time and thought exclusively to the problems they were addressing, thus actually attaining a level that truly was beyond comparison.

    Jaynes was a really smart guy, but no one can be a genius all the time. He did make at least one notable blunder in Bayesian probability theory -- a blunder he could have avoided if only he'd followed his own rules for careful probability analysis.

    You come across as very intelligent when you stick to your areas of expertise, like probability theory, AI and cognitive biases, but some of your more tangential stuff can seem a little naive. Compared to the other major poster on this blog, Robin, I'd say you come across as smarter but less "wise", if that means anything to you. I'm not even a huge fan of the notion of "wisdom", but if there's something you're missing, I think that's it.


    If you haven't read it, Simonton's Origins of Genius draws a nice distinction between mental agility and long-term intellectual significance, and explores the correlation between the two. Not a terribly well-written book, but certainly thought-provoking.

    @EY: We are the cards we are dealt, and intelligence is the unfairest of all those cards. More unfair than wealth or health or home country, unfairer than your happiness set-point. People have difficulty accepting that life can be that unfair, it's not a happy thought. "Intelligence isn't as important as X" is one way of turning away from the unfairness, refusing to deal with it, thinking a happier thought instead. It's a temptation, both to those dealt poor cards, and to those dealt good ones. Just as downplaying the importance of money is ... (read more)

    It's simply dissolving some cognitive illusions he shouldn't have had in the first place, but that most of us have probably had at some point in our lives. If you've got intelligence at 2 standard deviations above average, and you overestimate your own intelligence by one standard deviation (which is probably a pretty common mistake, and if anything underestimates the effect) than you'll see that you're probably the most intelligent person you interact with on a regular basis. If you're out at 3 standard deviations, it may not be until college that you see that some of your fellow students, or at least some of your professors, are indisputably smarter than you. If you're out at 4 or 5 standard deviations, as I imagine Eliezer is (I myself can't honestly peg myself past 3.5 standard deviations, which means I'm probably around 2 standard deviations above average and can't really distinguish beyond 2 standard deviations above my own level), I have some difficulty imagining what that must be like, only that even in the things you read you won't find many minds as formidable as your (perception of) your own, and even rarer will be minds that clearly surpass your own. But I think he is in the camp of trying to improve human intelligence (or at least human rationality, gwern seems to be the better poster child for improving human intelligence). Hence the sequences.
    Is a home-schooled person well positioned to judge that sort of thing? They're the smartest kind in a class of one.
    Not sure how homeschooling is relevant here, but speaking as a homeschooled person: it goes both ways, you're also the stupidest person in a class of one.
    Sidenote: I'd homeschool my kids if it were allowed where I live.
    (This seems like the wrong thread for a protracted discussion but I'm happy to say more in an open thread or via PM if you want to hear more, although it sounds like it's a moot point for you.)
    (I do want to hear more, go ahead using any means you'd like.)
    OK: http://lesswrong.com/r/discussion/lw/gbw/open_thread_january_1631_2013/8bdp

    Eliezer, I've been watching you with interest since 1996 due to your obvious intelligence and "altruism." From my background as a smart individual with over twenty years managing teams of Ph.D.s (and others with similar non-degreed qualifications) solving technical problems in the real world, you've always struck me as near but not at the top in terms of intelligence. Your "discoveries" and developmental trajectory fit easily within the bounds of my experience of myself and a few others of similar aptitudes, but your (sheltered) arrogance has always stood out. I wish you continued progress, not so much in ever-sharper analysis, but in ever more effective synthesis of the leading-edge subjects you pursue.

    How much do you worry about age 40? Is that just based on your father? Conway passed 40 before Marcello was born.

    If not, why aren't you in the camp of those who wish to improve human intelligence?

    I'll take this one because I'm almost certain Eliezer would answer the same way.

    Working on AI is a more effective way of increasing the intelligence of the space and matter around us than increasing human intelligence is. The probability of making substantial progress is higher.

    I disagree. Human intelligence is clearly misoptimised for many goals, and I see no clear evidence that it's easier to design a new intelligence from scratch than to optimise the human one. They have very different possible effects "FOOM!" vs. "We are awaiting GFDCA [Genetics, Food Drugs and Cybernetics Administration] approval of this new implant/chimerism/genehack", so the average impact of human-optimisation may be lower, but my probability estimate for human-improvement tech is much higher.

    Wow, chill out, Eliezer. You're probably among the top 10, certainly in the top 20, most-intelligent people I've met. That's good enough for anything you could want to do. You are ranked high enough that luck, money, and contacts will all be more important factors for you than some marginal increase in intelligence.

    First, same question as Douglas: what is it with the brick wall at 40?

    Second: This is another great post, its rare for people to expose their thoughts about theirselves in such an open way. Congratulations!

    Regarding your ability, I'm just a regular guy(studied Math in college) but your writings are the most inspiring I've ever read. So much self-reflection about intelligence and the thinking process. The insight about how certain mental processes feel is totally new to me. You have helped me a lot to identify my own blind spots and mistakes. Now I can look... (read more)

    I second Robin's comment.

    A friend of mine, Steve Jordan, once asked me just how smart I thought he and I were. I answered that I think that no-one is really as smart as the two of us both think we are. You see, for many many people it is possible to choose a weighting scheme among a dozen or so factors contribute to intellectual work such that they are the best. You simply define the vector to their point on the "efficient aptitude frontier" as "real intelligence". A dozen or so people associated with this blog and/or with SIAI and a smaller number who aren't appear to me to be on points of the "known to Michael Vassar efficient aptitude frontier", though not necessarily equally mission-critical points. For my "save the world dream team" I would pick a 25-year-old Steve Jobs over a 25-year-old Terrance Tao, though I'd like both of course.

    Manuel, "enroll in a grad program for AI" != "you're smart, you should go to college".

    Kragen, the short answer is, "It's easy to talk about the importance of effort if you happen to be Hamming." If you can make the ante for the high-stakes table, then you can talk about how little the ante counts for, and the importance of playing your cards well. But if you can't make the ante...

    Robin, it's not blind faith in math or math for the sake of impressiveness, but a specific sense that the specific next problems I have to solve, will require more math than I've used up to this point. Not Andrew J. Wiles math, but Jaynes doesn't use Wiles-math either. I quite share your prejudice against math for the sake of looking impressive, because that gets you the wrong math. (Formality isn't about Precision?)

    Ken, it's exclusively my work that gives me the motivation to keep working on something for years, but things like pride can give me the motivation to keep working on something for the next minute. I'll take whatever sources of motivation I can get (er, that aren't outright evil, of course).

    Douglas, yes, my father changed at 40. But one of my primary sources... (read more)

    Robin, it's not blind faith in math or math for the sake of impressiveness, but a specific sense that the specific next problems I have to solve, will require more math than I've used up to this point.

    I'm curious if this is still your sense, and if so, what kind of math are you talking about?

    My sense is that currently the main problems in FAI are philosophical. Skill in math is obviously very useful, but secondary to skill in philosophy, because most of the time it's still "I have no idea how to approach this problem" instead of "Oh, if I can just solve this math problem, everything will be clear".

    ...or I'm strictly dumber than Conway, dominated by him along all dimensions. Maybe, if I could find a young proto-Conway and tell them the basics, they would blaze right past me, solve the problems that have weighed on me for years, and zip off to places I can't follow.

    Marcello observed "In terms of philosophical intuition, you are head and shoulders above Conway." Making progress in FAI theory seems to require a combination of rationality, good philosophical intuition, math talent, motivation, and prerequisite background knowledge. (Am I leaving out anything?) Out of these, perhaps good philosophical intuition is rarest, in large part because we don't know how to teach it (or screen for it at a young age). Is this a problem you've considered?

    Philosophy? Really??? My impression of philosophy has been that it is entirely divorced from anything concrete or reality-based with no use in solving concrete, reality-based problems -- that all the famous works of philosophy are essentially elaborate versions of late-night college bull sessions, like irresistible forces vs. immovable objects, or trees falling in forests that do/don't make a sound. After working my way through a lot of the posts here, I now think that most of philosophy comes down to semantics and definitions of terms (i.e. Eliezer's excellent analysis of the tree-sound argument), and that what remains is still entirely divorced from reality and real-world uses. What have I missed? How does philosophy bring anything useful to the table?

    What have I missed? How does philosophy bring anything useful to the table?

    You appear to have missed philosophy. If you take a historical view, all of our contemporary subjects come from philosophy. The core of philosophy is precisely the sort of things we care about here - having an accurate picture of the world and understanding its true nature. To that end, ancient philosophers such as Aristotle invented logic, studied the natural world, discovered the inner workings of the human body, and started to investigate the laws that tie together everything in the world.

    Properly defining philosophy in current times is somewhat difficult based on this - now what was once called philosophy is instead called "Science" and other fields. So what is left is anything we don't already have an answer for. Philosophers are those who know what questions still need to be asked, and care about investigating them in a manner that will give them a more accurate picture of the world.

    To be a little more concrete (to give a specific example), the field of Ethics is considered a subfield of philosophy (largely because its questions are not yet settled), and one relevant question to FAI is simply "How should an AI behave?", which is an ethical question.

    I can agree that most of "science and other fields" came out of what was called "philosophy" if you go back far enough. It just seems that once you pull out all the "science and other fields" what is left has no use for solving practical problems -- including AI. Like the pages and pages of debate I've seen here on Less Wrong about "philosophical" stuff like the nature of morality, or free will, or "zombies" with no consciousness. Obviously a lot of people feel that discussing these topics is worthwhile, but I just don't see the use of it. In continuing to plod through the older writings here, I've seen numerous passages from Eliezer that disparage philosophy's usefulness, including these that I hit today: and So I'm still baffled by the comment here that currently the main problems in FAI are philosophical. Is there a summary or chain of posts that spells out this change in position? Or will it just gradually emerge if I manage to read all the posts between those quotes from 2008, up to the quote from 2011? Or, is this just your opinion, not Eliezer's?

    Note the distinction between those things being done in the field of "Philosophy", versus philosophy itself. Note that this:

    philosophy really is important, but it is only practiced effectively from within a science

    is an endorsement of philosophy itself, though the quote goes on to say that the way that much philosophy is done in academia is pretty useless. So I'm not seeing anything that should generate confusion. When Wei Dai said that the problems are philosophical, that does not entail that the problems should be solved by people with doctorates in Philosophy.

    Or, is this just your opinion, not Eliezer's?

    While I sometimes imagine myself one of the world's foremost experts on the writings of Eliezer, any non-quoted words are my own.

    It just seems that once you pull out all the "science and other fields" what is left has no use for solving practical problems -- including AI.

    If you think that logic, ethics, applied ontology, epistemology, and philosophy of mind all have no use in AI, then I think you will find yourself in a minority.

    4Wei Dai
    Hi Beth. I agree with Thom's answer, and you can also get a better sense of where I'm coming from if you read Metaphilosophical Mysteries and follow the first link in that post.

    As lukeprog has pointed out, almost everything Eliezer has written on LW is philosophy.

    I'd be curious to hear the answers to Wei's questions (in the sibling comment)...
    6Wei Dai
    From subsequent discussions, my understanding is that Eliezer doesn't think it's likely that he can recruit/train another FAI researcher with a level of philosophical competence similar to himself, and that he's planning/hoping to solve the main philosophical (as opposed to mathematical) problems himself. I've argued against this plan but I'm guessing that Eliezer is probably still set on this course. Others at SIAI may have more reservations about it.
    9Eliezer Yudkowsky
    I certainly intend to try that recruiting thing (Paul Christiano ain't half bad) but recruiting philosophy seems much less straightforward than recruiting mathematical talent. If I have to resolve it all myself, I wouldn't flinch from trying. It seems like that part should be less difficult in an absolute sense than the rest of the labor, though that might just be comparative advantage talking. The resolutions to philosophical confusions usually seem relatively straightforward once you have them, in my experience so far.
    5Wei Dai
    As I asked in the linked comment, if you're the only philosopher in the team, how will others catch your mistakes? I would not trust that when you stop feeling confused, the problem has been correctly solved, or that your feelings of confusion are a reliable indicator of problems existing in the first place. Having Paul involved certainly makes me feel better, and if you do succeed in recruiting more philosophy talent, then the issue may be moot. But I'm still concerned about your readiness to go it alone, and what that implies about your views, not only of how hard the problems are, but also how much one needs to worry about making philosophical mistakes.
    6Eliezer Yudkowsky
    Do you have some reliable way of recruiting? What's the policy alternative? You do what you gotta do, if ends up being just you, nonetheless, you do what you gotta do. Zero people won't make fewer mistakes than one person.
    6Wei Dai
    Quoting Carl Shulman from about a year ago: I'm not sure if he had both math and philosophy in mind when he wrote that or just math, but in any case surely the same principle applies to the philosophy. If you don't reach a high confidence that the philosophy behind some FAI design is correct, then you shouldn't move forward with that design, and if there is only one philosopher on the team, you just can't reach high confidence in the philosophy.
    6Eliezer Yudkowsky
    This does not sound correct to me. Resolutions of simple confusions usually look pretty obvious in retrospect. Or do you mean something broader by "philosophy" than trying to figure out free will?

    Did you read the rest of that thread where I talked about how in cryptography we often used formalizations of "security" that were discovered to be wrong years later, and that's despite having hundreds of people in the research community constantly trying to attack each other's ideas? I don't see how formalizing Friendliness could be not just easier and less error prone than formalizing security, but so much so that just one person is enough to solve all the problems with high confidence of correctness.

    Or do you mean something broader by "philosophy" than trying to figure out free will?

    I mean questions like your R1 and R2, your "nonperson predicate", how to distinguish between moral progress and moral error / value drift, anthropic reasoning / "reality fluid". Generally, all the problems that need to be solved for building an FAI besides the math and the programming.

    Yes, formalizing Friendliness is not the sort of thing you'd want one person doing. I agree. I don't consider that "philosophy", and it's the sort of thing other FAI team members would have to be able to check. We probably want at least one high-grade actual cryptographer.

    Of the others, the nonperson predicate and the moral-progress parts are the main ones where it'd be unusually hard to solve and then tell that it had been solved correctly. I would expect both of those to be factorable-out, though - that all or most of the solution could just be published outright. (Albeit recent experience with trolls makes me think that no insight enabling conscious simulations should ever be published; people would write suffering conscious simulations and run them just to show off... how confident they were that the consciousness theory was wrong, or something. I have a newfound understanding of the utter... do-anything-ness of trolls. This potentially makes it hard to publicly check some parts of the reasoning behind a nonperson predicate.) Anthropic reasoning / "reality fluid" is the sort of thing I'd expect to be really obvious in retrospect once solved. R1 and R2 should be both obvious in retrospect, and publishable.

    I have hopes that an upcoming post on the Lob Problem will offer a much more concrete picture of what some parts of the innards of FAI development and formalizing look like.

    Yes, formalizing Friendliness is not the sort of thing you'd want one person doing. I agree. I don't consider that "philosophy", and it's the sort of thing other FAI team members would have to be able to check.

    In principle, creating a formalization of Friendliness consists of two parts, conceptualizing Friendliness, and translating the concept into mathematical language. I'm using "philosophy" and "formalizing Friendliness" interchangeably to refer to both of these parts, whereas you seem to be using "philosophy" to refer to the former and "formalizing Friendliness" for the latter.

    I guess this is because you think you can do the first part, then hand off the second part to others. But in reality, constraints about what kinds of concepts can be expressed in math and what proof techniques are available means that you have to work from both ends at the same time, trying to jointly optimize for philosophical soundness and mathematical feasibility, so there is no clear boundary between "philosophy" and "formalizing".

    (I'm inferring this based on what happens in cryptography. The people creating new security concepts, the people writing down the mathematical formalizations, and the people doing the proofs are usually all the same, I think for the above reason.)

    8Eliezer Yudkowsky
    My experience to date has been a bit difference - the person asking the right question needs to be a high-grade philosopher, the people trying to answer it only need enough high-grade philosophy to understand-in-retrospect why that exact question is being asked. Answering can then potentially be done with either math talent or philosophy talent. The person asking the right question can be less good at doing clever advanced proofs but does need an extremely solid understanding of the math concepts they're using to state the kind-of-lemma they want. Basically, you need high math and high philosophy on both sides but there's room for S-class-math people who are A-class philosophers but not S-class-philosophers, being pointed in the right direction by S-class-philosophers who are A-class-math but not S-class-math. If you'll pardon the fuzzy terminology.
    What happened (if you don't mind sharing)?
    I get the impression that you have something different in mind as far as 'trolls' go than fools who create stereotypical conflicts on the internet. What kind of trolls are these?
    The kind who persuade depressed people to commit suicide. The kind who post people's address on the internet. The kind that burn the Koran in public.

    My psychological model says that all trolls are of that kind; some trolls just work harder than others. They all do damage in exchange for attention and the joy of seeing others upset, while exercising the limitless human ability to persuade themselves it's okay. If you make it possible for them to do damage on their home computers with no chance of being arrested and other people being visibly upset about it, a large number will opt to do so. The amount of suffering they create can be arbitrarily great, so long as they can talk themselves into believing it doesn't matter for and other people are being visibly upset to give them the attention-reward.

    4chan would have entire threads devoted to building worse hells. Yes. Seriously. They really would. And then they would instantiate those hells. So if you ever have an insight that constitutes incremental progress toward being able to run lots of small, stupid, suffering conscious agents on a home computer, shut up. And if somebody actually does it, don't be upset on the Internet.

    In case anyone doubts this, as a long-time observer of the 4chan memeplex, I concur.
    Related: How often does 4chan torture animals? That's pretty easy to pull off. Are they doing it all the time and I haven't noticed, or is there some additional force preventing it (e.g. Anonymous would hunt them down and post their details online, or 4chan all just like animals.)
    I remember that once, a Facebook page was hacked into (I guess) and started posting pictures and stories about tortured animals. Everybody went WTF and the page was shut down a few days later. I've never been there, but plenty of people on the internet do. Facebook pages against vivisection etc. seem to get way more likes than those in favour of it, the meme that humanity had better become extinct because wildlife would be better off is quite widespread, and some people even rejoice when a hunter dies (though this is a minority stance).
    Not often. Hurting animals is generally considered Not OK on 4chan, to the extent that anything is Not OK on 4chan. There are a few pictures and stories that get passed around (some kids kicking a cat against a wall like a football, shoveldog, etc), but many fewer than the human gore pictures. 4channers mostly aggregate this stuff from all over and post it to be edgy and drive people who aren't edgy enough away from 4chan. And yeah, to the extent that people do torture animals in current events (as opposed to past stories), vast hordes of moralfags and raiders from 4chan tend to hunt them down and ruin their lives.
    I wonder if this might happen to people running hells too? I lack the domain expertise to judge if this is ludicrous or impossible to predict or what.
    Really depends on whether the beings in the hell are cute and empathetic. Humans don't like to hurt things that are cute and empathetic, and don't like them getting hurt. Otherwise we don't care.

    4chan would have entire threads devoted to building worse hells. Yes. Seriously. They really would. And then they would instantiate those hells.

    They really would at that. It seems you are concerned here about malicious actual trolls specifically. I suppose if the technology and knowledge was disseminated to that degree (before something actually foomed) then that would be the most important threat. My first thoughts had gone towards researchers with the capabilities and interest to research this kind of technology themselves who are merely callous and who are indifferent to the suffering of their simulated conscious 'guinea pigs' for the aforementioned .

    So if you ever have an insight that constitutes incremental progress toward being able to run lots of small, stupid, suffering conscious agents on a home computer

    At what level of formalization does this kind of 'incremental progress' start to count? I ask because your philosophical essays on reductionism, consciousness and zombies is something that seems to be incremental progress towards that end (but which I certainly wouldn't consider a mistake to publish or a net risk).

    Related. (I'm not a huge fan of SCP in general, but I like a few stories with the "infohazard" tag, and I'm amused by how LW-ish those can get.) Eliezer could argue that the incremental progress towards stopping the risk outweighs the danger, same as with the general FAI/uFAI secrecy debate.
    I can't find the quote on that page. Is it from somewhere else (or an earlier version) or am I missing something?
    White text. (Apparently there's a few more hidden features in the entry, but I only found this one.)
    Ah, thanks.
    I, um, still can't find it. This white text is on the page you linked to, yes? About the videos that are probably soultraps? EDIT: Nevermind, got it.
    I think EY vastly overrates security through obscurity. Szilard keeping results about graphite and neutrons secret happened before the Internet; now there's this thing called the Streisand effect.
    "The Sims" is often heralded as the best-selling videogame of all time, and it attracts players of all ages, races and genders from all across the world and from all walks of life.[citation needed] Now imagine if the toons in the game could actually feel what was happening to them and react believably to their environment and situation and events? I'm sure I don't need to quote the Rules of Acquisition; everyone here should know where this leads if word of such a technique gets out.
    There have always been those who would pull the wings off flies, stomp on mice, or torture kittens. Setting roosters, fish, or dogs to fight each other to death remains a well-known spectacle in many rural parts of the world. In Shakespeare's day, Londoners enjoyed watching dogs slowly kill bulls or bears, or be killed by them; in France they set bushels of cats on fire to watch them burn. Public executions and tortures, gladiatorial combat among slaves, and other nonconsensual "blood sports" have been common in human history. What's the difference?
    How do you know that they don't?
    The average individual could not hold private gladiatorial contests, on a whim, at negligible cost. Killing a few innocents by torture, as public spectacle, is significantly less than repeatedly torturing large groups, as private entertainment, for as little as the average individual would have paid for their ticket to the cockfight. Also, some people reckon the suffering of animals doesn't matter. They're wrong, but they wouldn't care about most of your examples (or at least they would claim it's because they increase the risk you'll do the same to humans, which is a whole different kettle of fish.)
    Not to mention the sizeable fraction of car drives who will swerve in order to hit turtles. What the hell is wrong with my species?
    Link is broken. ... seriously? Poor turtles >:-(
    It was mentioned recently on Yvain's blog and a few months ago on LW (can't find it right now).
    Previous discussion of this on LW
    How do you know that they don't?

    Why do you always have to ask subtly hard questions? I can just see see your smug face, smiling that smug smile of yours with that slight tilt of the head as we squirm trying to rationalize something up quick.

    Here's my crack at it: They don't have what we currently think is the requisite code structure to "feel" in a meaningful way, but of course we are too confused to articulate the reasons much further.

    Thank you, I'm flattered. I have asked Eliezer the same question, not sure if anyone will reply. I hoped that there is a simple answer to this, related to the complexity of information processing in the substrate, like the brain or a computer, but I cannot seem to find any discussions online. Probably using wrong keywords.
    Information integration theory seems relevant.
    Not directly related. I think it has a lot to do with being roughly isomorphic to how a human thinks, which requires large complexity, but a particular complexity. When I evaluate such questions IRL, like in the case of helping out an injured bird, or feeding my cat, I notice that my decisions seem to depend on whether I feel empathy for the thing. That is, do my algorithms recognize it as a being, or as a thing. But then empathy can be hacked or faulty (see for example pictures of african children, cats and small animals, ugly disfigured people, far away people, etc), so I think of a sort of "abstract empathy" that is doing the job of recognizing morally valuable beings without all the bugs of my particular implementation of it. In other words, I think it's a matter of moral philosophy, not metaphysics.
    Well, I can't speak for the latest games, but I've personally read (some of) the core AI code for the toons in the first game of the series, and there was nothing in there that made a model of said code or attempted any form of what I'd even call "reasoning" throughout. No consciousness or meta-awareness. By being simulated by the code simulating the game in which they "are", they could to some extent be said to be "aware" of certain values like their hunger level, if you really want to stretch wide the concept of "awareness". However, there seems to be no consciousness anywhere to be 'aware' (in the anthropomorphized sense) of this. Since my priors are such that I consider it extremely unlikely that consciousness can exist without self-modeling and even more unlikely that consciousness is nonphysical, I conclude that there is a very low chance that they can be considered a "mind" with a consciousness that is aware of the pain and stimuli they receive. The overall system is also extremely simple, in relative terms, considering the kind of AI code that's normally discussed around these parts.
    Why would them feeling it help them "react believably to their environment and situation and events"? If they're dumb enough to "run lots of small, stupid, suffering conscious agents on a home computer", I mean. Of course, give Moore time and this objection will stop applying.
    We're already pretty close to making game characters have believable reactions, but only through clever scripting and a human deciding that situation X warrants reaction Y, and then applying mathematically-complicated patterns of light and prerecorded sounds onto the output devices of a computer. If we can successfully implement a system that has that-function-we-refer-to-when-we-say-"consciousness" and that-f-w-r-t-w-w-s-"really feel pain", then it seems an easy additional step to implement the kind of events triggering the latter function and the kind of outputs from the former function that would be believable and convincing to human players. I may be having faulty algorithmic intuitions here though.
    Well, if they were as smart as humans, sure. Even as smart as dogs, maybe. But if they're running lots of 'em on a home PC, then I must have been mistaken about how smart you have to be for consciousness.
    I used to torture my own characters to death a lot, back in the day. EDIT: Not to mention what I did when playing Roller Coaster Tycoon.
    The favourite Sim household of my housemate was based on "Buffy the Vampire Slayer". Complete with a graveyard constructed in the backyard. Through the judicial application of "remove ladder" from the swimming pool. And this is all without any particular malice!
    Most any incremental progress towards AGI, or even "just" EMs, would be dual use (if not centuple use) and could be (ab)used for helping achieve such enterta ... vile and nefarious purposes. In fact, it is hard to imagine realistic technological progress that can solely be used to run lots of small, stupid, suffering conscious agents but not as a stepping stone towards more noble pursuits (... such as automated poker playing agents).
    You know, I want to say you're completely and utterly wrong. I want to say that it's safe to at least release The Actual Explanation of Consciousness if and when you should solve such a thing. But, sadly, I know you're absolutely right re the existence of trolls which would make a point of using that to create suffering. Not just to get a reaction, but some would do it specifically to have a world they could torment beings. My model is not that all those trolls are identical (In that I've seen some that will explicitly unambiguously draw the line and recognize that egging on suicidal people is something that One Does Not Do, but I also know (seen) that all too many gleefully do do that.)
    It's worth noting that private torture chambers seem different to trolling, but a troll can still set up a torture chamber - they just care about people's reaction to it, not the torture itself.
    Wishing I could disagree with you, and, suspiciously, I find myself believeing that there would be enough vigilante justice to discourage hellmaking - after all, the trolls are doing it for the attention, and if that attention comes in the form of people posting your details and other people breaking into your house to steal your computer and/or murder you (for the greater good) then I doubt there will be many takers. I just wish I could trust that doubt.* *(Not expressing a wish for trust pills.) EDIT: Animal experimentation and factory farming are still popular, but they have financial incentive ... and I vaguely recall that some trolls kicked a dog across a football field or something and were punished by Anonymous. That's where the analogy comes from, anyway, so I'd be interested if someone knows more.
    I sometimes wonder if this does not already exist, except for the suffering and consciousness being merely simulated. That is, computer games in which the entire purpose is to inflict unspeakable acts on powerless NPCs, acts whose depiction in prose or pictures would be grossly illegal almost everywhere. But I've never heard of such a thing actually existing.
    What sort of acts are we talking here? Because I'm genuinely having trouble thinking of any "acts whose depiction in prose or pictures would be grossly illegal almost everywhere" except maybe pedophilia. Censorship and all that. And there are some fairly screwed-up games out there, although probably not as bad as they could be if designed with that in mind (as opposed to, y'know, the enjoyment of the player.) Well would you, if it was grossly illegal to describe the contents?
    I didn't want to be explicit, but you thought of the obvious example. For the sort of 4chan people Eliezer mentioned, these would be completely congruent. It is well known that illegal pornography exists on non-interactive media. For interactive media, all I've ever heard of is 18-rated sex scenes.
    I can't think of any other examples, though. ... maaaybe. Again, I'm not sure exactly what you have in mind. Good point. Indeed, it's well known that child porn exists on some level. In fact ... I do vaguely recall something about a Japanese game about rape causing a moral panic of some kind, so ... EDIT: In fact, it featured kids too! RapeLay. It's ... fairly horrible, although I think someone with the goal of pure horribleness would do .. better? Worse? Whatever.
    That stupid reason is, at core, nihilistic solipsism - and it's not as stupid as you'd think. I'm not saying it's right, but it does happen to be the one inescapable meme-trap of philosophy. To quote your own fic, their reason is "why not?" - and their consciousness was not grown such that your impassioned defense of compassion and consideration have any intrinsic factor in their utility function.
    At least for now, it'd take a pretty determined troll who could build an em for the sole purpose of being a terrible person. Not saying some humanity-first movement mightn't pull it off, but by that point you could hopefully have legal recognition (assuming there's no risk or accidental fooming and they pass the Turing test.)
    I don't think we're talking ems, we're talking conscious algorithms which aren't necessarily humanlike or even particularly intelligent. And as for the Turing Test, one oughtn't confuse consciousness with intelligence. A 6-year old human child couldn't pass off as an adult human, but we still believe the child to be conscious, and my own memories indicate that I indeed was at that age.
    Well, I think consciousness, intelligence and personhood are sliding scales anyway, so I may be imagining the output of a Nonperson Predicate somewhat differently to LW norm. OTOH, I guess it's not a priori impossible that a simple human-level AI could fit on something avvailable to the public, and such an insight would be ... risky, yeah. Upvoted.
    First of all, I also believe that consciousness is most probably a sliding scale. Secondly, again you just used "human-level" without specifying human-level at what, at intelligence or at consciousness; as such I'm not sure whether I actually communicated adequately my point that we're not discussing intelligence here, but just consciousness.
    Well, they do seem to be correlated in any case. However, I was referring to consciousness (whatever that is.)
    Re non-person predicates, do you even have a non-sharp (but non-trivial) lower bound for it? How do you know that the Sims from the namesake game aren't persons? How do we know that Watson is not suffering indescribably when losing a round of Jeopardy? And that imagining someone (whose behavior you can predict with high accuracy) suffering is not as bad as "actually" making someone suffer? If this bound has been definitively established, I'd appreciate a link.
    It's unclear where our intuitions on the subject come from or how they work, and they are heavily .... distorted ... by various beliefs and biases. OTOH, it seems unlikely that rocks are conscious and we just haven't extrapolated far enough to realize. It's also unclear whether personhood is binary or there's some kind of sliding scale. Nevertheless, it seems clear that a fly is not worth killing people over. Even a person who has never introspected about their moral beliefs can still know that murder is wrong. They're more likely to make mistakes, but still.
    How are these related? One is epistemology and one is ontology.
    1Wei Dai
    Can you give some more examples of this, besides "free will"? (I don't understand where your intuitions comes from that certain problems will turn out to have solutions that are obvious in retrospect, and that such feelings of obviousness are trustworthy. Maybe it would help me see your perspective if I got some more past examples.)
    A tree falls in a forest with no-one to hear it. Does it make a sound?
    I don't class that as a problem that is discussed by professional philosophers. It's more of a toy question that introduces the nature of phil. problems -- and the importance of asking "it depends on what you mean..." -- to laypeople.
    I agree, but that's not what I was aiming for. It's an example of obviousness after the fact, not philosophers being wrong/indecisive.
    It's not an example that lends much credence to the idea that all problems can be solved that way, even apart from the generalisation-from-one-example issue.
    I'm not claiming it proves anything, and I'm not taking sides in this discussion. Someone asked for an example of something - something which varies from person to person depending on whether they've dissolved the relevant confusions - and I provided what I thought was the best example. It is not intended to prove anyone's point; arguments are not soldiers.
    The counterargument to "arguments are not soldiers" is "a point should have a point".
    It wasn't an argument at all. That you chose to interpret it as an enemy soldier is your mistake, not mine. It's not a weak soldier, it's a ... medic or something.
    And the other example being generalised from isnt that good
    Do you have an example in mind where a certain philosophical question claimed to have been solved or dissolved by Eliezer turned out to be not solved after all, or the solution was wrong?
    6Wei Dai
    * Shut Up and Divide? * Beware Selective Nihilism Also, instances where Eliezer didn't seem to realize that a problem existed until someone pointed it out to him: * Marcello on responses to moral arguments being possibly order-dependent or subject to butterfly effects * Nesov's Counterfactual Mugging * Me arguing that Solomonoff Induction may not be adequate * Eliezer seemingly not realizing that making certain kinds of physics "unimaginable" is a bad idea
    9Eliezer Yudkowsky
    Order-dependence and butterfly effects - knew about this and had it in mind when I wrote CEV, I think it should be in the text. Counterfactual Mugging - check, I don't think I was calling TDT a complete solution before then but the Counterfactual Mugging was a class of possibilities I hadn't considered. (It does seem related to Parfit's Hitchhiker which I knew was a problem.) Solomonoff Induction - again, I think you may be overestimating how much weight I put on that in the first place. It's not a workable AI answer for at least two obvious reasons I'm pretty sure I knew about from almost-day-one, (a) it's uncomputable and (b) it can't handle utility functions over the environment. However, your particular contributions about halting-oracles-shouldn't-be-unimaginable did indeed influence me in toward my current notion of second-order logical natural induction over possible models of axioms in which you could be embedded. Albeit I stand by my old reply that Solomonoff Induction would encompass any computable predictions or learning you could do about halting oracles in the environment. (The problem of porting yourself onto any environmental object is something I already knew AIXI would fail at.)
    5Wei Dai
    Ok, I checked the CEV writeup and you did mention these briefly. But that makes me unsure why you claimed to have solved metaethics. What should you do if your FAI comes back and says that your EV shows no coherence due to order dependence and butterfly effects (assuming it's not some kind of implementation error)? If you're not sure the answer is "nothing", and you don't have another answer, doesn't that mean your solution (about the meaning of "should") is at least incomplete, and possibly wrong? You said that TDT solves Parfit's Hitchhiker, so I don't know if you would have kept looking for more problems related to Parfit's Hitchhiker and eventually come upon Counterfactual Mugging. Both of these can be solved without also solving halting-oracles-shouldn't-be-unimaginable. For (a), solve logical uncertainty. For (b), switch to UDT-with-world-programs. Also, here is another problem that maybe you weren't already aware of.
    Wouldn't that kind of make moral reasoning impossible?
    You never did any engineering-level mathematical modeling of real system, did you? The main difficulty is not proving the theorems, it is finding the right axioms to describe the relevant aspects of the system and the properties of interest. And that's where errors often occur. Now, typical engineering tasks pale in comparison to the task you are trying to undertake: creting a fully specified mathematical model of ethics. Most likely it's just the Dunning–Kruger effect Just like when you "resolved" the interpretation of quantum mechanics? Well, good thing that you are never going to make anything close to an AGI and that AGI risk is probably overrated, otherwise it wouldn't end well...
    For the record: I, too, want an FAI team in which Eliezer isn't the only one with Eliezer-level philosophical ability or better. This is tougher than "merely" finding 1-in-10-million math talents, but still do-able. What am I doing about it? I wrote a post encouraging a specific kind of philosophical education that I think will be more likely to produce Eliezer-level philosophers than a "normal" philosophical education (or even a CMU or UPitts one). When Louie came up with the idea to write a list of Course recommendations for Friendliness researchers, I encouraged it. Also, one of the reasons I ended up supporting the plan to launch CFAR in 2012 was its potential not only to make people more effective at achieving their goals, but also to learn ways to make some people better philosophers (see my last paragraph here). And there's more, but I can't talk about it yet. Also, as Eliezer said, Paul Christiano's existence is encouraging.
    What about Kawoomba's existence? :-( CFAR and related are good efforts at raising the sanity waterline (which is an average), not so much for identifying the extreme outliers that could Alan-Turing their way towards an FAI. Those will make waves on their own. Such grassroots organisations may be good ways of capturing the attention of a wider audience, although second to publishing in the field / personally building a network at conferences. The time horizon and viability of having a few hundred self-selected college aged students and trying to grow them into a seminal figure of extraordinary capabilities seems prohibitive, especially when there are already exceedingly capable people at Stanford et al, who already bring the oomph and just lack the FAI-motivation.
    Can you name some older academics that have the requisite philosophical skill? (And if your first line isn't a joke, perhaps you can link me to some of your own philosophical works?)
    Sipser, Russell&Norvig et al are core parts of your proposed philosophical curriculum, Louis' course recommendations reads like my former grad CS reading list. It follows that, say, many with or pursuing a PhD in Machine Learning and related have also picked up a majority of your desired (per your recommendations) philosophical skills. I'm not postulating that Bayesian superstars also make the best drummers and fencing masters, but between your analytical CS-style philosophy and Machine Learning groups there is a cross-domain synergy effect that comes with the clarity of designing minds - or advanced algorithms. (As for myself, the first line was meant as a joke - alas! How sad!)
    No, I wouldn't say that. The problem is that we (humans) don't know how to teach the philosophical skill I'm talking about, so there aren't classes on it, so I can only recommend courses on "the basics" or "prerequisites." I don't know how to turn a math/CS PhD under Stuart Russell into the next Eliezer Yudkowsky.
    The only thing that's certain is that somebody has trouble properly apprehending your intelligence.

    Vassar - your English is encrypted - more an assumption of intelligence than a sign.

    EY - I admire your work. Along with Robin this is the best Show in Town and I will miss it, when it stops.

    I actually doubt whether you are accomplishing anything - but this does not seem so important to me, because the effort itself is worthwhile. And we are educated along the way.

    This is a youthful blog with youthful worries. From the vantage point of age worrying about intelligence seems like a waste of time and unanswerable to boot.

    But those are the stones in your shoes.

    @Jef Allbright:

    Can you be concrete and specific about where Eliezer is or has been arrogant?

    "Most intelligent people I've met" is not informative, we need to give quantitative estimates. My estimate is calibrated based on knowing people who passed various screenings, such as math, physics and programming contests (including at international level), test results on screening exams to top universities, performance in hard university courses, people starting to grasp research and programming, etc. Based on population of regions covered by various screenings, and taking age, gender and different background into account, I can approximately rate these people on the "1 in XXX" scale. I'd say that you need to be at a level of 1 in 300 or so to be able to deeply understand any technical field of human knowledge given reasonable effort, and 1 in 100 to be a competent technical specialist. There is a significant difference (which can cash out as, say, 3x speedup at obtaining given level of aptitude) between people who are 1 in 1000 and 1 in 10000. I know too few people beyond 1 in 10000 (about top 30 in a contest over population of 20 million within a 3-year age interval, given average lifespan of 60 and background selection of 1 in 3 top people to enter the con... (read more)

    My own potential intelligence does worry me fairly often. I am currently studying to become an engineer and hope to work on some of the awesome ideas I read about on sites like this. The thing is though, I wasted the first twenty third years of my life. I am currently at twenty-five years old and I have been forced to pretty much start from scratch on everything from social skills to education and after two years I think I am making some headway. I am even starting to understand what Eliezer talks about in all these posts and apply it to my own life as bes... (read more)

    I am in a situation that is practically the same as yours. As an ex-child prodigy, I identify with the insecurities this thread is about. However, having studied Math at Colege for a year and a half, and having met there people who were so beyond my level that it heavily obstructed communication with them, that they were so far beyond my understanding they could have not been human for all it mattered, I can confidently say that my ambitions of genius-dom have been utterly crushed beyond repair. As I am, I will be very content to be a competent, respected engineer. This said, I think Eleizer is protesting too much. You shouldn't care how smart you are, only that you are getting your job done: that arrogance of yours needs some serious work, as you know, we know you know, etc. And it seems unfortunate in that regard that exposure to a large amount of far more brilliant minds than yours is an experience that you will have to go seriously out of your way to get, since that is probably the shortest, easiest way to humility. If you find an alternate path that is not as bitter, please let us know.

    Let me give a shout out to my 1:50 peeps! I can't even summarize what EY has notably accomplished beyond highlighting how much more likely he is to accomplish something. All I really want is for Google to stop returning pages that are obviously unhelpful to me, or for a machine to disentangle how the genetic code works, or a system that can give absolute top notch medical advice, or something better than the bumbling jackasses[choose any] that manage to make policy in our country. Give me one of those things and you will be one in a million, baby.


    I suppose you could google "(arrogant OR arrogance OR modesty) eliezer yudkowsky" and have plenty to digest. Note that the arrogance at issue is neither dishonest nor unwarranted, but it is an impairment, and a consequence of trade-offs which, from within a broader context, probably wouldn't be taken in the same way.

    That's as far as I'm willing to entertain this line of inquiry, which ostensibly neutral request for facts appears to belie an undercurrent of offense.

    You're probably among the top 10, certainly in the top 20, most-intelligent people I've met. That's good enough for anything you could want to do.

    Okay, I realize you're going to read that and say, "It's obviously not good enough for things requiring superhuman intelligence!"

    I meant that, if you compare your attributes to those of other humans, and you sort those attributes, with the one that presents you the most trouble in attaining your goal at the top, intelligence will not be near the top of that list for you, for any goal.

    @Jef Allbright:

    I suppose you could google "(arrogant OR arrogance OR modesty) eliezer yudkowsky" and have plenty to digest.

    Well, I was asking you, not google. But it seems that you are not willing to stand behind your words, making claims then failing to provide evidence when asked. Refering to a third party is an evasive maneuver. Show us your cards!

    That's as far as I'm willing to entertain this line of inquiry, which ostensibly neutral request for facts appears to belie an undercurrent of offense.

    That's your supposition.

    Eliezer, can you clarify what you mean by "You'll note that I don't try to modestly say anything like, "Well, I may not be as brilliant as Jaynes or Conway, but that doesn't mean I can't do important things in my chosen field."

    Because I do know... that's not how it works."

    He didn't reply to this, so I'll take a stab at unpacking/justifying that statement. For the number of people in the world, smaller numbers are more conservative, so suppose that number is 6 billion. For the number of truly distinct things that there are to be good at, larger numbers are more conservative, so let us suppose that number to be 600,000. (I doubt if any reader can come up with even 10,000 such things, even allowing trivial variations like "experimental high energy physicist with subspecialty A" and "experimental high energy physicist with subspecialty B".) Under those conservative assumptions, it is mathematically necessary that AT LEAST 99% of people are not in the top 100 at anything. Further, that boundary only happens with the additional conservative assumptions that 1) literally everyone pursues the exact career (among 600,000 choices!) that an infallible oracle told them they were the best at, and 2) people match careers in such a way that each one is assigned by the oracle to at least 100 people. The farther reality is from those, the more the percentage gets closer to 100%, or the rank required to be at the top of something dips lower than 100. Because of the preceding argument, it seems likely that most people are not very important to the field to which they are the most important. Therefore, it seems irrational to believe "I am important to my chosen field" without specific, relatively strong evidence that this is the case. Certainly "lots of people say that that's true of everyone" isn't strong enough, since that evidence has the more-plausible alternate explanation that people say that because it's comforting to them to believe it, in ignorance of whether it's actually true.

    Vladimir Nesov: thanks for your comment. I found it insightful.

    You say 'That's not how it works.' But I think that IS how it works!

    If progress were only ever made by people as smart as E.T. Jaynes, humanity would never have gotten anywhere. Even with fat tails, intelligence is still roughly normally distributed, and there just aren't that many 6 sigma events. The vast majority of scientific progress is incremental, notwithstanding that it's only the revolutionary achievements that are salient.

    The real question is, do you want Friendly A.I. to be achieved? Or do you just want friendly A.I. to be achieved by YOU? There'... (read more)

    I find myself, except in the case of people with obvious impairments, completely unable to determine how intelligent someone is by interacting with them. Sometimes I can determine who is capable of performing specific tasks, but I have little confidence in my ability to assess "general intelligence".

    To some extent, this is because different people have acquired different skills. Archimedes of Syracuse may have been the greatest mathematician in history, but he wouldn't be able to pass the exams in a high school calculus class. Obviously, the reas... (read more)


    I believe that you don't really understand something until you can explain it to someone else, and have them understand it, too.

    There's basically two reasons to get called arrogant. One is acting like you're better when you aren't. The other is refusing to politely pretend that the inferential chasm is small. Given where E is and where the mass of humanity are, if I had to make blind-guess assignments for 100 accusers picked at random, and I assigned them all into the "inferential distance" bin, I don't think I'd be wrong once. So, a person asking to be put, or to put some accuser into the "undeserved airs" bin, had better show some sharp evidence!

    Not so fast: "Major mathematical advances past age fifty"
    Why was this down voted? The comic is a take on a fairly prevalent belief, heck, Hardy said it! I wish more people on this forum would explain why they were down voting something, that on the face of it, seems reasonable. I'm up voting this. EDIT: When I posted this, I was of the opinion that the comic was just giving a funny take on the maths is a young man's game thing. Now, after looking it several times, I am of the opinion that it was trying to poke fun of this said misconception. And, giving the benefit of doubt to the original poster, I still stand by my upvote.
    Let's find an archaeologist to exhume the remains of the long since dead reader who downvoted that comment a mere 52 months ago. Who knows what his thought process was? Did he hunt saber-tooths and not appreciate the cave-man like quality of XKCD stick figures? And where did he even get his computer, or did he MacGyver a Turing Machine out of sticks and stones?
    Turns out he's still alive!
    I did. I didn't want to explain why because it's a long conversation I don't want to have, but basically I think this quote promotes a misleading conception of mathematics. For the record, Hardy is thought to have been suffering from depression when he wrote this.
    Am I misreading the comic? It appears to be making fun of that quote. Maybe it's more haha only serious than I thought? EDIT: By the way, I upvoted you even though I think you may have misinterpreted the comic in question, as I am in favor of people explaining their downvotes. Keep up the good work!
    I know, and I've read too, that Hardy was apparently not in the pink when he said this. And in all honesty the comic seems to be making fun of the conception that maths was for the young.
    Me too. (Not sure that a link to that xkcd comic in this context “on the face of it, seems reasonable”, though.)
    Does a link to a comic have a place in this forum? I don't know the answer to that, perhaps it is not. That said, my comment was more a reaction to other down votes, this just happened to be the straw I was commenting on.
    (I meant “context” not “comment” BTW -- fixed that.) Yes. I've posted such links myself. But that particular one seems to me to have very dubious topicality.
    Really? I think, after staring at it for some time, that the comic is making fun of the thinking that maths is a young man's game.

    "Perhaps it is the fear of being too late that is causing you distress. Perhaps you fear that humanity is going to be destroyed because you didn't build an FAI soon enough. Perhaps you fear that your life will end some 10,000 years sooner than you'd like."

    Humanity's alleged demise is not the only possible way he could be too late. I wonder where Eliezer would turn his attention if someone (or some group) solved the problems of FAI before him.

    Eliezer has written a number of times about how comparing your intelligence and rationality to those aroun... (read more)

    Eliezer: It seems to me that uncertainty about your abilities is dwarfed by uncertainty about the difficulty of the problem.

    Doug S: The median college graduate in a technical field probably would test in the 95th percentile on most IQ tests and at the 98th percentile on tests weighted heavily towards non-vocabulary crystalline g

    Eliezer: Not sure to what extent this helps or answers your questions, but I increasingly as of late find that much of my current "cached wisdom" seems to be derived from stuff you've said.

    As far as as actually finding the next generation or whatever, maybe some people here that know how ought to start some "private school for the gifted" that explicitly is meant to try to act almost like a Bayes Dojo or whatever and otherwise train up people in really precise thinking?

    If such an insitution is ever built, and it is what it claims to be (remember that all causes want to become cults), then, believe me, I'd send my children there as a top priority. Such a school should not be for geniuses only, but for the development of the teaching of the Methods of Rationality to all of humanity. To set an example to be followed by all who wish not to be left behind. I remeber Aldous Huxley mentioning, in Brave New World Revisited, a similar project by some philantrope, who wished to make children immune to propaganda and manipulation. The authorities shut it down because, they said, it turned the teenagers and young adults too cynical and disobedient to Authority. For example, they were able to calmly analyze the Drill Sergeant's speech as it was given. And obviously, we couldn't have that, could we? I think such a project has considerably greater chances of success nowadays.

    While Conway has a huge jump on you in mathematical ability, and I'm pretty sure you're not going to catch up to him, rest assured that you are not strictly dumber than Conway in every respect.

    You should bear in mind how the statement "Maybe anything more than around one standard deviation above you starts to blur together, though that's just a cool-sounding wild guess" might apply to me. If your guess is literally true, then, because math is my strong-suit, high mathematical ability is the smartest kind of smart that I can detect at all. For m... (read more)

    Eliezer: Look on the bright side, you haven't yet relegated yourself to being a mere administrator and occasional sounding board for others' AI research projects! Ego subjugation is a bitch, but it can have minor rewards of self-satisfaction when actions driven by pressure-free buckshot mental synthesis actually bear fruit. I don't envy that it's of no help to you that the luxury of being carefree relies on the knowledge that smarter people are doing the heavy lifting, and today you're at the top tier of that brain chain!

    Maksym: We actually do need someone to translate all this OB stuff very badly, though maybe it's desirable to wait for the book. Still, someone should be presenting it. As for convincing smart college students, there are three fairly separate barriers here, those to rationality, those of information and those to action. I recommend working on barriers to rationality and action first and in conjunction, belief second, and let people find the info themselves. Politics is the natural subject to frame as rationality. Simply turn every conversation where ... (read more)


    Dude, you honestly make me ill sometimes. You spoke nothing of the circumstances that got these people to where they are or where they came from. There are people just as "sparkly" and some smarter than these people who have not had the opportunity that these people have. You are blinded by your arrogance and are locked in the present time. You are a smart guy, but you would have a lot to gain in building interpersonal wisdom.

    The sparkle you describe is meaningless; non-sparkling borderline-autistic types do just as fine work as the most invigoratingly sparkling individuals. I choose to sparkle through my work, in quiet solitude, not through swaying my limbs excitedly, motor-mouthing like a sports commentator on amphs.


    Its a benefit for me to read this post having not read your others, because I can give you an untainted view of it. You are too concerned with intelligence. As long as you stay in this state, you are unusable, and pass up opportunities on becoming usable.
    Snap out of it. Accept that there are more intelligent people than you, and they are not flailing, they just get on with it.

    Again, I have difficulty understanding why so many people place such a high value on 'intelligence' for its own sake, as opposed to a means to an end. If Eliezer is worried that he does not have enough mathematical intelligence to save the universe from someone else's misdesigned AI, than this is indeed a problem for him, but only because the universe will not be saved. If someone else saves the universe instead, Eliezer should not mind, and should go back to writing sci-fi novels. Why should Eliezer's ego cry at the thought of being upstaged? He shoul... (read more)

    Of course I want there to be someone smarter than me to take over, from an altruistic perspective. Or even from just a selfish perspective of being scared, wanting a vacation, and feeling a bit isolated.

    And of course if that actually happened, it would be a severe blow to my ego.

    And so long as I can do the expected-utility-maximizing thing and invest the appropriate amount of effort into preparing for the possibility without betting the whole farm on it, I have no intention of hacking at my emotions on either score.

    I know how you feel, in a couple ways. My high-school guidance counselor looked at my middle school transcript and told me I might realistically aspire to go to a UC school (as opposed to a school in the Cal State system). (I ended up going to Harvard and Caltech.) On the other hand, the year I finished my Ph.D. (at the age of 29) one of my college acquaintances, a brilliant mathematician, became one of the youngest full professors in the history of Princeton University, and when my Ph.D. advisor was 29 he had already been a professor at Caltech for sev... (read more)

    It's a general rule, I've observed, that you can't discriminate between levels too far above your own.
    Although I agree with this in general, it seems that there are a few specific counterexamples. For example, it seems that people with very low ability in sports actually can discriminate ability from the local high school level to the international stage.

    Do other people agree? If so, what do you propose distinguishes between intelligence/mathematical ability and athletic ability?

    If so, what do you propose distinguishes between intelligence/mathematical ability and athletic ability?
    To evaluate athletic ability, you use your judgment. What can you use to evaluate your judgment?

    It is possible for a person to produce an accurate evaluation of a subset of their own intellectual skills, but certain skills cannot be evaluated, because presumptions about those skills are required for the evaluation to take place. You should not ask questions about subjects in which you presume you already know the answers, and you cannot ask questions about subjects where answers must be presumed in order to be able to ask at all.

    Lara, I don't think they value it "for its own sake" as opposed to as a means to an end; rather, they see it as a necessary condition for achieving their ends, and are worried they don't have what it takes. Nothing but an anxiety trip.

    And of course, there's also the ego thing -- when people build superiority over others into their self-image. This is counterproductive, of course. When someone else demonstrates that they're "smarter" than you by offering unexpected insight, you don't fatalistically wallow in jealous misery; you listen to... (read more)

    Eliezer, Komponisto,

    I understand the anxiety issues of, 'Do I have what it takes to accomplish this..."

    I don't understand why the existence of someone else who can would damage Eliezer's ego. I can observe that many other people's sense of self is violated if they find out that someone else is better at something they thought they were the best at-- the football champion at HS losing their position at college, etc. However, in order for this to occur, the person needs to 1) in fact misjudge their relative superiority to others, and 2) value the supe... (read more)

    I have no idea if it's a natural human quality. It's surely one of my qualities. It's not that I would permit my mind to think verbal thoughts like "How good it is to be above others." But there's a zest in being the best. It feels good to complete a difficult race and it feels good to win a gold medal; they are separate, different good feelings. I can imagine people who would only care about having completed the challenge, but they wouldn't be me.

    Since my mind doesn't want whatever I choose it to want, I accept that both desires are a part ... (read more)

    I can imagine people who would only care about having completed the challenge, but they wouldn't be me.

    I'm not sure there are any people like this who are capable of occasionally winning. OTOH, the prospect of never winning might force someone to rationalize themselves into this position.


    The proof is in the math and/or in the protopudding, is it not? There are people/groups who already have either or both. If you have neither, what's your sense of relative achievement/skill/IQ based on?

    What (math &/ prototype) do you have? If none, what do you plan to have, when? It seems you'd have to blaze past those who already have their stuff out in the real world behaving ever more AGI-ishly by the day, to meet your criteria for success. A tall order to be sure.

    Some else wrote

    This is a youthful blog with youthful worries. From the vantage point of age worrying about intelligence seems like a waste of time and unanswerable to boot.

    and I find this observation insightful, and even a bit understated.

    Increasingly, as one ages, one worries more about what one DOES, rather than about abstract characterizations of one's capability.

    Obviously, one reason these sorts of questions about comparative general intelligence are unanswerable is that "general intelligence" is not really a rigorously def... (read more)

    Achieving great things seems always to be a mixture of general intelligence, specialized intelligence, wise choice of the right problems to work on, and personality properties like persistence ...

    With a pinch of being in the right place and the right time, bake on 350 for 10-30 years.

    I kind of disagree with you. First, what we call "general intelligence" is itself a form of specialized intelligence: specializing optimizing successful outcomes in real time in our apparent reality. so the mix you recommend in "achieving great things" would itself be "general intelligence", not general intelligence plus something else (other than luck).

    Since most people who "achieve great things" seem to me to be playing life at least in part as a poker game (they don't seem to put all their cards out on the ta... (read more)

    Increasingly, as one ages, one worries more about what one DOES, rather than about abstract characterizations of one's capability.

    This definitely happened to me. Between the ages of about 10 - 14, I was utterly obsessed with finding out what my IQ was. Somehow, somewhere along the way, I'd picked up the notion that Smartness in quantity was the most important thing a person could possibly have.

    And it drove me frankly batty not knowing how much Smartness I had, because (a) I was insecure and felt like I needed to find out I had a "high enough" ... (read more)

    Eliezer, don't think to yourself that you only have until you are 40. As somebody else noted and you didn't acknowledge, Marcello was not yet born when Conway passed 40. You mentioned your father, and I don't know the specifics, but surely you know that plenty of people have done great work, sometimes their best, past 40, and that with every passing year, due to advances in health, medicine, etc., "youth" extends further and further into our life.

    And as another poster mentioned, I have almost no doubt that Von Neumann would have blown Einstein (p... (read more)

    Supporting Ben Goertzel's comment:

    Michael Shermer revised his book, Why People Believe Weird Things, to contain a chapter called “Why Smart People Believe Weird Things”. In it, he quotes studies by Hudson, Getzels, and Jackson showing that “creativity and intelligence are relatively orthogonal (i.e., unrelated statistically) at high levels of intelligence. Intuitively, it seems like the more intelligent people are the more creative they will be. In fact, in almost any profession significantly affected by intelligence, once you are at a certain level ... (read more)