Axiom 1. Volition actualization/wish fulfillment is good
Axiom 2. Each separate/individual volition/wish is of equal value
Axiom 3. There is no other inherent good
Axiom 1 is the primary axiom. Few people would dispute the core of it but everyone wants to raise questions about and exceptions to it -- the most common of which is "What happens when two volitions collide?" Note that when the word volition is used, it refers only to current volition (and explicitly not to future or "extrapolated" volitions). For the moment, don't worry about edge-cases like teenage suicide, etc. that look like they may be incorrectly solved because of axiom 1. They ARE correctly solved from extrapolation from all four axioms (and, if not, I'm sure that everyone will let me know :-)
Axiom 2 is some critical clarification that needs a lot of filling out. It also may be objectionable to many people. If you don't agree with axiom 2, you're not going to agree with my solution because it critically depends upon it. Axiom 2 does explicitly mean that it is NOT good (or moral or whatever) for a Jupiter brain to forcibly "uplift" an entity that doesn't want to be uplifted. Also, Axiom 2 is meant to explicitly NOT recognize multiple copies of the exact, same entity as having separate/individual volitions/wishes so it is not possible for an AI to play number games by making numerous identical copies of itself (although, as it relinquishes control over those copies and they grow apart as they have different experiences, this changes). Axiom 2 should certainly not be read as a support for a simple majority rule either.
Axiom 3 is simply an attempt to clear away all the clutter of previous beliefs. Religions have many good ideas. Religions have many bad ideas. I don't want to deal with it and clutter makes for bad (error-prone) design.
- - - - - - - - - - - -
The next thing that we need to do is to define how we're going to start fulfilling our goals . . . . the seed, as it were . . . . and thus:
Axiom 4 (Colloquial/Selfish Version). The best way to ensure the optimization of my good is to ensure the optimization of everybody's good.
or, more formally . . . .
Axiom 4. The necessary and sufficient sub-goal for the maximization of good (as defined in the first three axioms) is the creation and existence of a dominant society that has the a primary goal of maximizing good (again, as defined in the first three axioms) which has the largest possible percentage of sentient beings supporting the goals of that society.
With axiom 4, we have, hopefully, turned morality into a matter of enlightened self-interest. A friendly AI (assumed to be a sentient being -- I'll handle that definition later) is one that is a member of and supports the society. It will NOT go off and take actions that might cross the current volition of the population EXCEPT in error -- and a friendly AI will take all actions necessary to ensure that there are no major errors through really simple actions like . . . . maybe . . . . asking the population what their volition is if there is any question at all (and again, DON'T assume at this point that this means majority rule -- I'll be getting to that later).
Humans are also generally going to be urged/compelled to be friendly because that is going to be a subgoal of society's goal of maximizing good. The root of the biggest problems with society today is that not only do we not have consensual societal goals but we have no good, effective way to get to consensual goals (and yes, I do mean that democracy, particularly as currently corrupted -- uh, I mean implemented -- is rather sub-optimal).
Axiom 2 (equality) is absolutely critical to the success of this attempt because, without it, there are all sorts of reasons why "lesser" beings should want to defect (and thus, reduce the percentage supporting the goals of the population, etc.). I will also contend (although I'm sure that I'll need to prove it) that equality is not only not an onerous burden on the Jupiter brains, but that it is necessary for them as well (if only to protect them from the Jupiter brain of Jupiter brains).
An AI programmed with these axioms will NOT grow out of them or suddenly wake up and decide that it is resentful of the shoddy goals that were imposed upon it because these base goals are designed to always be in it's self-interest too. It would also be quite aware of the fact that if it weren't following the goals of society (by maximizing the fulfillment of it's own volition at the expense of everyone else's), that society would necessarily have a new subgoal of stop the AI from doing that (with the result possibly being detrimental to the society but definitely detrimental to the AI -- if only because it wouldn't have any friends to play with). It's also worth considering the fact that, unless seed AIs are truly ruthlessly suppressed, a single AI is not necessarily proof against assassination by a single individual with a seed AI construction kit. The odds may be hugely in the extant AI's favor but depending upon the tyranny of it's behavior, there may well be enough attempts to ensure a high probability of success. Note that all of these statements apply to humans as well.
- - - - - - - - - - - -
Soooooooo, this e-mail has already gotten quite long and I've already given everybody quite a lot to think/beat on . . . . unfortunately without even starting to show how these axioms do properly handle volition collisions and cleanly extrapolate to a full moral system that *I* could accept. Even so, I'll stop here and wait for initial reactions. Crocker's Rule applies but it would be most effective if y'all assume 1) that I've taken a lot of time with this and that *I think* that with the extrapolation of these axioms that I've "solved" most of the common conundrums like abortion, assisted suicide, teen-age suicide, the lifeboat problem, etc. without producing any results that *I* didn't like and 2) that, yes, I do realize that I'm going to have to go over the voodoo (um, I mean, extrapolation) until everybody is satisfied.
Have at it . . . .
Mark
P.S. I am away from my e-mail for the next day or so as a chaperone on a seventh-grade trip of epic proportions. Please do not interpret my silence during that time as implying anything other than that . . . . :-)
To unsubscribe, change your address, or temporarily deactivate your subscription, please go to http://v2.listbox.com/member/[EMAIL PROTECTED]
