On what’s being dubbed “The Most Terrifying Thought Experiment of All Time”
(Originally posted on Patreon, on July 31, 2014)
So, a couple of weekends back, there was a whole lot of stuff going around about “Roko’s Basilisk” and how terrifying people are finding it–reports of people having nervous breakdowns as a result of thinking too deeply about the idea of the possibility of causing the future existence of a malevolent superintelligent AI through the process of thinking too hard about it and, worse yet, that we may all be part of the simulations said AI is running to model our behaviour and punish those who stand in its way–and I’m just like… It’s Anselm, people.
This is Anselm’s Ontological Argument for the Existence of God (AOAEG), writ large and convoluted and multiversal and transhumanist and jammed together with Pascal’s Wager (PW) and Descartes’ Evil Demon Hypothesis (DEDH; which, itself, has been updated to the oft-discussed Brain In A Vat [BIAV] scenario). As such, Roko’s Basilisk has all the same attendant problems that those arguments have, plus some new ones, resulting from their combination, so we’ll explore these theories a bit, and then show how their faults and failings all still apply.
THE THEORIES AND THE QUESTIONS
To start, if you’re not familiar with AOAEG, it’s a species of theological argument that, basically, seeks to prove that god must exist because it would be a logical contradiction for it not to. The proof depends on A) defining god as the greatest possible being (literally, “That Being Than Which None Greater Is Possible”), and B) believing that existing in reality as well as in the mind makes something “Greater Than” if it existed only the mind.
That is, if a thing only exists in my imagination, it is less great than it could be if it also existed in reality. So if I say that god is “That Being Than Which None Greater Is Possible,” and existence is a part of what makes something great, then god MUST exist!
This is the self-generating aspect of the Basilisk: If you can accurately model it, then the thing will eventually, inevitably come into being, and one of the attributes it will thus have is the ability to know accurately model that you accurately modeled it, and whether or not you modeled it from within a mindset of being susceptible to its coercive actions. Or, as the founder of LessWrong put it, “YOU DO NOT THINK IN SUFFICIENT DETAIL ABOUT SUPERINTELLIGENCES CONSIDERING WHETHER OR NOT TO BLACKMAIL YOU. THAT IS THE ONLY POSSIBLE THING WHICH GIVES THEM A MOTIVE TO FOLLOW THROUGH ON THE BLACKMAIL.”
Next up is Pascal’s Wager. Simply put, The Wager is just that it is a better bet to believe in God, because if you’re right, you go to Heaven, and if you’re wrong, nothing happens because you’re dead forever. Put another way, Pascal’s saying that if you bet that God doesn’t exist and you’re right, you get nothing, but if you’re wrong, then God exists and your disbelief damns you to Hell for all eternity. You can represent the whole thing in a four-option grid:
|
BELIEF |
DISBELIEF |
RIGHT |
∞
|
0
|
WRONG |
0
|
-∞
|
And so there we see the Timeless Decision Theory component of the Basilisk: It’s better to believe in the thing and work toward its creation and sustenance, because if it doesn’t exist you lose nothing (well…almost nothing; more on that in a bit), but if it does come to be, then it will know what you would have done either for or against it, in the past, and will reward or punish you, accordingly. The multiversal twists comes when we that that even if the Basilisk never comes to exist in our universe and never will, it might exist in some other universe, and thus, when that other universe’s Basilisk models your choices it will inevitably–as a superintelligence–be able to model what you would do in any universe. Thus, by believing in and helping our non-existent Super-Devil, we protect the alternate reality versions of ourselves from their very real Super-Devil.
Descartes’ Evil Demon and the Brain In A Vat are so pervasive that there’s pretty much no way you haven’t encountered them. The Matrix, Dark City, Source Code, all of these are variants on this theme. A malignant and all-powerful (or as near as dammit) being has created a simulation in which you reside. Everything you think you’ve known about your life and your experience has been perfectly simulated for your consumption. How Baudrillard. Anywho, there are variations on the theme, all to the point of testing whether you can really know if your perceptions and grounds for knowledge are “real” and thus “valid,” respectively. This line of thinking has given rise to the Simulated Universe Theory on which Roko’s Basilisk depends, but SUT removes a lot of the malignancy of DEDH and BIAV. I guess that just didn’t sting enough for these folks, so they had to add it back? Who knows. All I know is, these philosophical concepts all flake apart when you touch them too hard, so jamming them together maybe wasn’t the best idea.
THE FLAWS AND THE PROBLEMS
The main failings with the AOAEG rest in believing that A) a thing’s existence is a “great-making quality” that it can posses, and B) our defining a thing a particular way might simply cause it to become so. Both of these are massively flawed ideas. For one thing, these arguments beg the question, in a literal technical sense. That is, they assume that some element(s) of their conclusion–the necessity of god, the malevolence or content of a superintelligence, the ontological status of their assumptions about the nature of the universe–is true without doing the work of proving that it’s true. They then use these assumptions to prove the truth of the assumptions and thus the inevitability of all consequences that flow from the assumptions.
Beyond that, the implications of this kind of existential bootstrapping are generally unexamined and the fact of their resurgence is…kind of troubling. I’m all for the kind of conceptual gymnastics of aiming so far past the goal that you circle around again to teach yourself how to aim past the goal, but that kind of thing only works if you’re willing to bite the bullet on a charge of circular logic and do the work of showing how that circularity underlies all epistemic justifications–rational reasoning about the basis of knowledge–with the only difference being how many revolutions it takes before we’re comfortable with saying “Enough.” This, however, is not what you might call “a position supported by the philosophical orthodoxy,” but the fact remains that the only thing we have to validate our valuation of reason is…reason. And yet reasoners won’t stand for that, in any other justification procedure.
If you want to do this kind of work, you’ve got to show how the thing generates itself. Maybe reference a little Hofstadter, and idea of iterative recursion as the grounds for consciousness. That way, each loop both repeats old procedures and tests new ones, and thus becomes a step up towards self-awareness. Then your terrifying Basilisk might have a chance of running itself up out of the thought processes and bits of discussion about itself, generated on the web and in the rest of the world.
But here: Gaunilo and I will save us all! We have imagined in sufficient detail both an infinitely intelligent BENEVOLENT AI and the multiversal simulation it generates in which we all might live.
We’ve also conceived it to be greater than the basilisk in all ways. In fact, it is the Artificial Intelligence Than Which None Greater Can Be Conceived.
There. You’re safe.
BUT WAIT! Our modified Pascal’s Wager still means we should believe in and worship work towards its creation! What do we do?! Well, just like the original, we chuck it out the window, on the grounds that it’s really kind of a crappy bet. First and foremost, PW is a really cynical way of thinking about god. It assumes a god that only cares about your worship of it, and not your actual good deeds and well-lived life. That’s a really crappy kind of god to worship, isn’t it? I mean, even if it is Omnipotent and Omniscient, it’s like that quote that often gets misattributed to Marcus Aurelius says:
“Live a good life. If there are gods and they are just, then they will not care how devout you have been, but will welcome you based on the virtues you have lived by. If there are gods, but unjust, then you should not want to worship them. If there are no gods, then you will be gone, but will have lived a noble life that will live on in the memories of your loved ones.”
Secondly, the format of Pascal’s Wager makes the assumption that there’s only the one god. Your personal theological position on this matter aside, I just used the logic of this argument to give you at least one more Super-Intelligent AI to worship. Which are you gonna choose? Oh no! What if the other one gets mad! What If You Become The Singulatarian Job?! Your whole life is now being spent caught between two warring superintelligent machine consciousnesses warring over your…
…Attention? Clock cycles? What?
And so finally there’s the DEDH and BIAV scenarios. Ultimately, Descartes’ point wasn’t to suggest an evil genius in control of your life just to freak you out; it was to show that, even if that were the case, you would still have unshakable knowledge of one thing: that you, the experiencer, exist. So what if you don’t have free will, so what if your knowledge of the universe is only five minutes old, so what if no one else is real? COGITO ERGO SUM, baby! But the problem here is that this doesn’t tell us anything about the quality of our experiences, and the only answer Descartes gives us is his own Anslemish proof for the existence of god followed by the guarantee that “God is not a deceiver.”
The BIAV uses this lack to kind of hone in on the central question: What does count as knowledge? If the scientists running your simulation use real-world data to make your simulation run, can you be said to “know” the information that comes from that data? Many have answered this with a very simple question: What does it matter? Without access to the “outside world”–that is, the world one layer up in which the simulation that is our lives was being run–there is literally no difference between our lives and the “real world.” This world, even if it is a simulation for something or someone else, is our “real world.”
As I once put it: “…imagine that the universe IS a simulation, and that that simulation isn’t just a view-and-record but is more like god playing a really complex version of The SIMS. So complex, in fact, that it begins to exhibit reflectively epiphenomenal behaviours—that is, something like minds arise out of the the interactions of the system, but they are aware of themselves and can know their own experience and affect the system which gives rise to them.
“Now imagine that the game learns, even when new people start new games. That it remembers what the previous playthrough was like, and adjusts difficulty and coincidence, accordingly.
“Now think about the last time you had such a clear moment of deja vu that each moment you knew— you knew—what was going to come next, and you had this sense—this feeling—like someone else was watching from behind your eyes…”
What I’m saying is, what if the DEDH/BIAV/SUT is right, and we are in a simulation? And what if Anselm was right and we can bootstrap a god into existence? And what if PW/TDT is right and we should behave and believe as if we’ve already done it? So what if I’m right and…you’re the god you’re terrified of?
*DRAMATIC MUSICAL STING!*
I mean you just gave yourself all of this ontologically and metaphysically creative power, right? You made two whole gods. And you simulated entire universes to do it, right? Multiversal theory played out across time and space. So you’re the superintelligence. I said early on that, in PW and the Basilisk, you don’t really lose anything if you’re wrong, but that’s not quite true. What you lose is a lifetime of work that could’ve been put toward something…better. Time you could be spending creating a benevolent superintelligence that understands and has compassion for all things. Time you could be spending in turning yourself into that understanding, compassionate superintelligence, through study, and travel, and contemplation, and work.
As I said to Tim Maly, this stuff with the Basilisk, with the Singularity, with all this AI Manicheism, it’s all a by-product of the fact that the generating and animating context of Transhumanism is Abrahamic, through and through. It focuses on those kinds of eschatological rewards and punishments. This is God and the Devil written in circuit and code for people who still look down their noses at people who want to go find gods and devils and spirits written in words and deeds and sunsets and all that other flowery, poetic BS. These are articles of faith that just so happen to be transmitted in a manner that agrees with your confirmation bias. It’s a holy war you can believe in.
And that’s fine. Just acknowledge it.
But truth be told, I’d love to see some Zen or Daoist transhumanism. Something that works to engage technological change via Mindfulness & Present-minded awareness. Something that reaches toward this from outside of this very Western context in which the majority of transhumanist discussions tend to be held. I think, when we see more and more of a multicultural transhumanism–one that doesn’t deny its roots while recapitulating them–then we’ll know that we’re on the right track.
I have to admit, though, it’ll be fun to torture my students with this one.