If the Singularity is the Rapture of the Nerds, self-modification of the brain must be Hell; a way to screw up to an arbitrary degree that most people don't even understand well enough to fear.
Phil: The first source I found was here: link
"The rationale for not divulging the AI-box method is that someone suffering from hindsight bias would say "I never would have fallen for that", when in fact they would." -Nick Tarleton
I also call it "reasoning by exception" since most of the people I know have studied more code than biases.
I tried the AI Box experiment with a friend recently. We called the result a tie of sorts, as the AI (me) got out of the original box in exchange for being subject to a bunch of restrictions set by the Gatekeeper, to be kept by verifiably modifying and publishing its own source code, so stringent that they were like a different sort of box.
Thomas: As I understand the quote, we do not perceive them. The machinery does. Then we are the thoughts it thinks about those thoughts.
Is it possible to get this filed under Humor so that I can view the series without having to hunt this entry down individually?
michael vassar: I meant "horrible" from my current perspective, much like I would view that future me as psychopathic and immoral. (It wouldn't, or if it did, it would consider them meaningless labels.)
Dynamically Linked: I'm using my real name and I think I'd do things that I (and most of the people I know) currently consider immoral. I'm not sure about using "admit" to describe it, thought, as I don't consider it a dark secret. I have a certain utility function which has a negative valuation of a hypothetical future self without the same utility function. While my current utility function has an entry for "truth", that entry isn't valued above all the others that Eliezer suggests disproving the way I understand it. But then, I'm still a bit confused on how the question should be read.
The post says "when you finally got hungry [...] what would you do after you were done eating?", which I take to understand that I still have desire and reason to eat. But it also asks me to imagine a proof that all utilities are zero, which confuses me because when I'm hungry, I expect a form of utility (not being hungry, which is better than being hungry) from eating. I'm probably confused on this point in some manner, though, so I'll try to answer the question the way I understand it, which is that the more abstracted/cultural/etc utilities are removed. (Feel free to enlighten/flame me on this point.)
I expect that I'd probably do a number of things that I currently avoid, most of which would probably be clustered under "psychopathy". I think there's something wrong with them now, but I wouldn't think that there was something wrong with them post-proof. Most of my behavior would probably stay the same due to enlightened self-interest, and I'm not sure what would change. For example, the child on the train tracks. My current moral system says I should pull them off, no argument. If you ripped that system away, I'd weigh off the possible benefit the child might bring me in the future (since it's in my vicinity, it's probably a First World kid with a better than average chance of a good education and a productive life) against considerations like overpopulation. I'd cheat on my Significant Other if I thought it would increase my expected happiness (roughly: "if I can get away with it"). I'd go on reading Overcoming Bias and being rational because rationality seems like a better tool for deciding what to eat when hungry, such as at the basic level of bread vs. candles, and generalise from there. (If that goes away, I probably die horribly from misnourishment.)
Something's been bugging me about MWI and scenarios like this: am I performing some sort of act of quantum altruism by not getting frozen since that means that "I" will be experiencing not getting frozen while some other me, or rather set of world-branches of me, will experience getting frozen?