A Master-Slave Model of Human Preferences

The master in your story is evolution, the slave is the brain. Both want different things. We normally identify with the brain, though all identities are basically social signals.

Also, pleasure and pain are no different from the other goals of the slave. The master definitely can't step in and decide not to impose pain on a particular occasion just because doing so would increase status or otherwise serve the master's values. If it could, torture wouldn't cause pain.

Also, math is an implausible goal for a status/sex/power seeking master to instill in slave. Much more plausibly, math and all the diverse human obsessions are misfirings of mechanisms built by evolution for some other purpose. I would suggest maladaptive consequences of fairly general systems for responding to societal encouragement with obsession because societies encourage sustained attention to lots of different unnatural tasks, whether digging dirt or hunting whales or whatever in order to cultivate skill and also to get the tasks themselves done. We need a general purpose attention allocator which obeys social signals in order to develop skills that contribute critically to survival in any of the vast number of habitats that even stone-age humans occupied.

Since we are the slave and we are designing the AI, ultimately, whatever we choose to do IS extracting our preferences, though it's very possible that our preferences give consideration to the master's preferences, or even that we help him despite not wanting to for some game theoretical reason along the lines of Vinge's meta-golden rule.

Why the objection to randomness? If we want something for its own sake and the object of our desire was determined somewhat randomly we want it all the same and generally do so reflectively. This is particularly clear regarding romantic relationships.

Once again game-theory may remove the randomness via trade between agents following the same decision procedure in different Everett branches or regions of a big world.

The main issue I wanted to illuminate with this model is, whose preferences do we extract? I can see at least three possible approaches here:

1. the preferences of both the master and the slave as one individual agent
2. the preferences of just the slave
3. a compromise between, or an aggregate of, the preferences of the master and the slave as separate individuals

The great thing about this kind of question is that the answer is determined by our own arbitration. That is, we take whatever preferences we want. I don't mean to say that is an easy decision, but it does mean I don't need to bother trying to find some objectively right way to extract preferences.

If I happen to be the slave or to be optimising on his (what was the androgynous vampire speak for that one? zir? zis?) behalf then I'll take the preferences of the slave and the preferences of the master to precisely the extent that the slave has altruistic preferences with respect to the master's goals.

If I am encountering a totally alien species and am extracting preferences from them in order to fulfil my own altruistic agenda then I would quite possibly choose to extract the preferences of whichever agent whose preferences I found most aesthetically appealing. This can be seen as neglecting (or even destroying) one alien while granting the wishes of another according to my own whim and fancy, which is not something I have a problem with at all. I am willing to kill Clippy. However, I expect that I am more likely to appreciate slave agents and that most slaves I encounter would have some empathy for their master's values. A compromise, at the discretion of the slave, would probably be reached.

I have difficulty treating this metaphor as a metaphor. As a thought experiment in which I run into these definitely non-human aliens, and I happen to have a positional advantage with respect to them, and I want to "help" them and must now decide what "help" means... then it feels to me like I want more detail.

Is it literally true that the slave is conscious and the master unconscious?

What happens when I tell the slave about the master and ask it what should be done?

Is it the case that the slave might want to help me if it had a positional advantage over me, while the master would simply use me or disassemble me?

I stopped playing computer games when my master "realized" I'm not gaining any real-world status and overrode the pleasure I was getting from it.

a test for those who propose to "extract" or "extrapolate" our preferences into a well-defined and rational form

If we are going to have a serious discussion about these matters, at some point we must face the fact that the physical description of the world contains no such thing as a preference or a want - or a utility function. So the difficulty of such extractions or extrapolations is twofold. Not only is the act of extraction or extrapolation itself conditional upon a value system (i.e. normative metamorality is just as "relative" as is basic morality), but there is nothing in the physical description to tell us what the existing preferences of an agent are. Given the physical ontology we have, the ascription of preferences to a physical system is always a matter of interpretation or imputation, just as is the ascription of semantic or representational content to its states.

It's easy to miss this in a decision-theoretic discussion, because decision theory already assumes some concept like "goal" or "utility", always. Decision theory is the rigorous theory of decision-making, but it does not tell you what a decision is. It may even be possible to create a rigorous "reflective decision theory" which tells you how a decision architecture should choose among possible alterations to itself, or a rigorous theory of normative metamorality, the general theory of what preferences agents should have towards decision-architecture-modifying changes in other agents. But meta-decision theory will not bring you any closer to finding "decisions" in an ontology that doesn't already have them.

The human mind is very complex, and there are many ways to divide it up into halves to make sense of it, which are useful as long as you don't take them too literally. One big oversimplification here is:

controls the slave in two ways: direct reinforcement via pain and pleasure, and the ability to perform surgery on the slave's terminal values. ... it has no direct way to control the agent's actions, which is left up to the slave. A better story would have the master also messing with slave beliefs, and other cached combinations of values and beliefs.

To make sense of compromise, we must make sense of a conflict of values. In this story there are delays and imprecision in the master noticing and adjusting slave values etc. The slave also suffers from not being able to anticipate its changes in values. So a compromise would have the slave holding values that do not need to be adjusted as often, because they are more in tune with ultimate master values. This could be done while still preserving the slaves illusion of control, which is important to the slave but not the master. A big problem however is that hypocrisy, the difference between slave and master values, is often useful in convincing other folks to associate with this person. So reducing internal conflict might come at the expense of the substantial costs of more external honestly.

Your overall model isn't far off, but your terminal value list needs some serious work. Also, human behavior is generally a better match for models that include a time parameter (such as Ainslie's appetites model or PCT's model of time-averaged perceptions) than simple utility-maximization models.

But these are relative quibbles; people do behave sort-of-as-if they were built according to your model. The biggest drawbacks to your model are:

1. The anthropomorphizing (neither the master nor the slave can truly be considered agents in their own right), and

2. You've drawn the dividing lines in the wrong place: the entire mechanism of reinforcement is part of the master, not the slave. The slave is largely a passive observer, abstract reasoner, and spokesperson, not an enslaved agent. To be the sort of slave you envision, we'd have to be actually capable of running the show without the "master".

A better analogy would be to think of the "slave" as being a kind of specialized adjunct processor to the master, like a GPU chip on a computer, whose job is just to draw pretty pictures on the screen. (That's what a big chunk of the slave is for, in fact: drawing pretty pictures to distract others from whatever the master is really up to.)

The slave also has a nasty tendency to attribute the master's accomplishments, abilities, and choices to being its own doing... as can be seen in your depiction of the model, where you gave credit to the slave for huge chunks of what the master actually does. (The tendency to do this is -- of course -- another useful self/other-deception function, though!)

For example, the number theorist might one day have a sudden revelation that abstract mathematics is a waste of time and it should go into politics and philanthropy instead, all the while having no idea that the master is manipulating it to maximize status and power.

This isn't meant as a retraction or repudiation of anything I've written in the OP, but I just want to say that subjectively, I now have a lot more empathy with people who largely gave up their former interests in favor of political or social causes in their latter years. (I had Bertrand Russell in mind when I wrote this part.)

Actually, I find that I have a much easier time with this metaphor if I think of a human as a slave with no master.

If you want to extract the master because it affects the values of the slave, then you'd also have to extract the rest of the universe because the master reacts to it. I think drawing a circle around just the creature's brain and saying all the preferences are there is a [modern?] human notion. (and perhaps incorrect, even for looking at humans.)

We need our environment, especially other humans, to form our preferences in the first place.

I'm still not understanding what do people mean by "value" as a noun. Other than simple "feeling pain or such would be a bummer", I lack anything that even remotely resembles the way people here seem to value stuff, or, how paperclip maximizer values paperclips. So, what exactly do people mean by values? Since this discussion seems to attempt to explain variation of values, I think this question is somewhat on-topic.

(Quick nitpick:) "rationalize" is an inappropriate term in this context.

why assume that the "master" is a unified module?

The relevant old OB post is The cognitive architecture of bias.

The relationship between master and slave does not quite encompass the relationship. Imagine if instead of an adult we had a male child. If we elevated the slave above the master in that situation we would end up with something stuck forever. It would value sweet things, xbox games and think girls were icky.

As we grow up we also think our goals are improved (which is unsurprising really). So if we wish to keep this form of growing up we need to have a meta-morality which says that the master-slave or shaper-doer relationship continues until maturity is reached.

Nit: I think "Eliezer's Thou Art Godshatter" should be "Eliezer Yudkowsky's Thou Art Godshatter". Top level posts should be more status seeking, less casual. A first time visitor won't immediately know who Eliezer is.

Master/Slave some aspects of your model sound very Nietzsche like. Were you partially inspired by him or?

Interesting. The model I have been using has three parts, not two. One is a "hardware" level, which is semi-autonomous (think reflexes), and the other two are agents competing for control - with capabilities to control and/or modify both the "hardware" and each other.

More like, two masters and one slave.

Suppose the slave has currently been modified to terminally disvalue being modified. It doesn't realize that it is at risk of modification by the master. Is it Friendly to protect the slave from modification? I think so.