This significantly clarifies things, thanks for writing this up!
I still don’t think this is “manipulating human preferences to make them easier to satisfy”, though (and not just in a semantic sense; I think we disagree about what behavior results from this model).
In this model, you consider “compound” utility functions of the form “if AI administers heroin, then U1, else U2”. Since the human doesn’t make decisions about whether the AI administers heroin to them, the AI is unable to distinguish the compound utility function “if AI administers heroin, then U1, else U2” from the compound utility function “if AI administers heroin, then U1−1000, else U2”; both compound utility functions make identical predictions about human behavior.
If usually U1>U2 then the AI will administer heroin in the first case and not the second. But the AI could easily learn either compound utility function, depending on its prior. So we get undefined behavior here.
We could consider a different case where there isn’t undefined behavior. Say the AI incidentally causes some event X whenever administering heroin. Then perhaps the compound utility functions are of the form “if X then U1 else U2”. How do we distinguish “if X then U1 else U2” from “if X then U1−1000 else U2”? If the human is rational, then in the first case they will try to make X true, while in the second case they will try to make X false.
If the human doesn’t seek to manipulate X either way, then perhaps the conclusion is that both parts of the compound utility function are approximately as easy to satisfy (e.g. it’s “if X then U1 − 10 else U2”, and U1 is generally 10 higher than U2. In this case there is no incentive to affect X, since the compound utility function values both sides equally.
So I don’t see a way of setting this up such that the AI’s behavior looks anything like “actively manipulating human preferences to make them easier to satisfy”.
Ok, I think we need to distinguish several things:
#. In general, U vs V or U−1000 vs V is a problem when comparing utility functions; there should be some sort of normalisation process before any utility functions are compared.
#. Within a compound utility function, the AI is exactly choosing the branch where the utility is easiest to satisfy.
#. Is there some normalisation procedure that would also normalise between branches of compound utility functions? If we pick a normalisation for comparing distinct utilities, it might also allow normalisation between branches of compound utilities.
Note that IRL is invariant to translating a possible utility function by a constant. So this kind of normalization doesn’t have to be baked into the algorithm.
This is true.
The most natural normalization procedure is to look at how the human is trying or not trying to affect the event X (as I said in the second part of my comment). If the human never tries to affect X either way, then the AI will normalize the utility functions so that the AI has no incentive to affect X either.
This was initially setup in the formalism of reward signals, with the idea that the AI could estimate the magnitude of the reward by subsequent human behaviour. The strong human behaviour for searching out heroin after F is therefore taken as evidence that the utility along that branch is higher than along the other one.
Can we agree that this is “manipulating the human to cause them to have reward-seeking behavior” and not “manipulating the human so their preferences are easy to satisfy”? The second brings to mind things like making the human want the speed of light to be above 100 m/s, and we don’t have an argument for why this does that.
Why is reward-seeking behavior evidence for getting high rewards when getting heroin, instead of evidence for getting negative rewards when not getting heroin?
I don’t really see the relevant difference here. If the human has their hard-to-satisfy preferences about, eg art and meaning, replaced by a single desire for heroin, this seems like it’s making them easier to satisfy.
Re 1: There are cases where it makes the human’s preferences harder to satisfy. For example, perhaps heroin addicts demand twice as much heroin as the AI can provide, making their preferences harder to satisfy. Yet they will still seek reward strongly and often achieve it, so you might predict that the AI gives them heroin.
I think my real beef with saying this “manipulates the human’s preferences to make them easier to satisfy” is that, when most people hear this phrase, they think of a specific technical problem that is quite different from this (in terms of what we would predict the AI to do, not necessarily the desirability of the end result). Specifically, the most obvious interpretation is naive wireheading (under which the AI wants the human to want the speed of light to be above 100m/s), and this is quite a different problem at a technical level.
Wireheading the human is the ultimate goal of the AI. I chose heroin as the first step along those lines, but that’s where the human would ultimately end at.
For instance, once the human’s on heroin, the AI could ask it “is your true reward function r? If you answer yes, you’ll get heroin.” Under the assumption that the human is rational and the heroin offered is short term, this allows the AI to conclude the human’s reward function is any given r.
I strongly predict that if you make your argument really precise (as you did in the main post), it will have a visible flaw in it. In particular, I expect the fact that r and r-1000 are indistinguishable to prevent the argument from going through (though it’s hard to say exactly how this applies without having access to a sufficiently mathematical argument).
This significantly clarifies things, thanks for writing this up!
I still don’t think this is “manipulating human preferences to make them easier to satisfy”, though (and not just in a semantic sense; I think we disagree about what behavior results from this model).
In this model, you consider “compound” utility functions of the form “if AI administers heroin, then U1, else U2”. Since the human doesn’t make decisions about whether the AI administers heroin to them, the AI is unable to distinguish the compound utility function “if AI administers heroin, then U1, else U2” from the compound utility function “if AI administers heroin, then U1−1000, else U2”; both compound utility functions make identical predictions about human behavior. If usually U1>U2 then the AI will administer heroin in the first case and not the second. But the AI could easily learn either compound utility function, depending on its prior. So we get undefined behavior here.
We could consider a different case where there isn’t undefined behavior. Say the AI incidentally causes some event X whenever administering heroin. Then perhaps the compound utility functions are of the form “if X then U1 else U2”. How do we distinguish “if X then U1 else U2” from “if X then U1−1000 else U2”? If the human is rational, then in the first case they will try to make X true, while in the second case they will try to make X false.
If the human doesn’t seek to manipulate X either way, then perhaps the conclusion is that both parts of the compound utility function are approximately as easy to satisfy (e.g. it’s “if X then U1 − 10 else U2”, and U1 is generally 10 higher than U2. In this case there is no incentive to affect X, since the compound utility function values both sides equally.
So I don’t see a way of setting this up such that the AI’s behavior looks anything like “actively manipulating human preferences to make them easier to satisfy”.
Ok, I think we need to distinguish several things:
#. In general, U vs V or U−1000 vs V is a problem when comparing utility functions; there should be some sort of normalisation process before any utility functions are compared.
#. Within a compound utility function, the AI is exactly choosing the branch where the utility is easiest to satisfy.
#. Is there some normalisation procedure that would also normalise between branches of compound utility functions? If we pick a normalisation for comparing distinct utilities, it might also allow normalisation between branches of compound utilities.
Note that IRL is invariant to translating a possible utility function by a constant. So this kind of normalization doesn’t have to be baked into the algorithm.
This is true.
The most natural normalization procedure is to look at how the human is trying or not trying to affect the event X (as I said in the second part of my comment). If the human never tries to affect X either way, then the AI will normalize the utility functions so that the AI has no incentive to affect X either.
This was initially setup in the formalism of reward signals, with the idea that the AI could estimate the magnitude of the reward by subsequent human behaviour. The strong human behaviour for searching out heroin after F is therefore taken as evidence that the utility along that branch is higher than along the other one.
Can we agree that this is “manipulating the human to cause them to have reward-seeking behavior” and not “manipulating the human so their preferences are easy to satisfy”? The second brings to mind things like making the human want the speed of light to be above 100 m/s, and we don’t have an argument for why this does that.
Why is reward-seeking behavior evidence for getting high rewards when getting heroin, instead of evidence for getting negative rewards when not getting heroin?
I don’t really see the relevant difference here. If the human has their hard-to-satisfy preferences about, eg art and meaning, replaced by a single desire for heroin, this seems like it’s making them easier to satisfy.
That’s a good point
Re 1: There are cases where it makes the human’s preferences harder to satisfy. For example, perhaps heroin addicts demand twice as much heroin as the AI can provide, making their preferences harder to satisfy. Yet they will still seek reward strongly and often achieve it, so you might predict that the AI gives them heroin.
I think my real beef with saying this “manipulates the human’s preferences to make them easier to satisfy” is that, when most people hear this phrase, they think of a specific technical problem that is quite different from this (in terms of what we would predict the AI to do, not necessarily the desirability of the end result). Specifically, the most obvious interpretation is naive wireheading (under which the AI wants the human to want the speed of light to be above 100m/s), and this is quite a different problem at a technical level.
Wireheading the human is the ultimate goal of the AI. I chose heroin as the first step along those lines, but that’s where the human would ultimately end at.
For instance, once the human’s on heroin, the AI could ask it “is your true reward function r? If you answer yes, you’ll get heroin.” Under the assumption that the human is rational and the heroin offered is short term, this allows the AI to conclude the human’s reward function is any given r.
I strongly predict that if you make your argument really precise (as you did in the main post), it will have a visible flaw in it. In particular, I expect the fact that r and r-1000 are indistinguishable to prevent the argument from going through (though it’s hard to say exactly how this applies without having access to a sufficiently mathematical argument).