If CEV produces whatever people value, do you think it would produce the above because you have different values than other people… ?
Yes. And thank you for phrasing it that way so I understand that is at least one explanation for my concern.
It seems beyond likely to me that the CEV you get will depend heavily on just who you include in your definition of “humans” whos volition must be considered in defining CEV. Even if CEV were intended to be just that subset of volitions that “everybody” would agree on (if they were smart enough), will your definition of everybody include paranoid schizophrenics? People born with severely deformed brains? Sociopaths? Republicans? The French? My point being that our intuition is of a “common definition of human we can all agree on” but the reality of 7 billion live humans plus a few billion easy to anticipate might have a non-intuitively large variation across its volition.
So if CEV includes a “veto power” in its definition granted to all humans defined broadly enough to include sociopaths, we lose many of the values that allow us to work cooperatively.
Further concerning me, I think it is likely that humanity benefits from a diversity in values. At one level, societies with different values have different levels of success under different challenges, and in something like survival of the fittest, the societies that thrive have values that work better than those that don’t. At another level, within a society diversity in values serves the group: the nurturers are caretakers, the nerds technologists, the sociopaths become leaders and work in security.
CEV as I have heard it described sounds like a core of values, a kernel that all FAI operating systems would have to include. It doesn’t sound like a set of values or a core of meta-values that would somehow incorporate in a single kernel all the variation in values that has served humanity so well.
So yes, I am concerned that CEV is impossible, but perhaps not provably impossible, that any actual attempts to build a CEV will have more to do with the values of the people building CEV rather than some undefinable generalization of humanity.
Another concern: AI with a CEV constraint will necessarily be less adaptable than AI with no CEV constraint. So in the absence of complete totalitarian control over where AI can come from, non-CEV AI once created would eventually out-compete CEV-based AI anyway, and all that effort would have been for naught.
Finally, what I think of as a Kurzweilian paradigm of AI makes more sense to me than the idea of independent AIs that exist separately from humans. Kurzweil seems to me talks more of enhancing existing humans, building on modules, interfacing us better, and so on. Eventually, perhaps, the enhanced human is 99% enhancement and 1% human and so it becomes a matter of attitude whether you still think of it as human. Do you think CEV is something that applies to building up enhanced humans (instead of independent entities)?
Also, it seems to me that avoiding a new technology (CEV) specifically because it will make your life too easy has a lot in common with living in a false world which is centuries out of date.
The ultimate technology that makes my life too easy is wireheading. That was just fun to say, I don’t actually recognize it as a great response to your point, but I throw it out there because there might be more to it than I am allowing.
I suppose one man’s dystopia is another woman’s brave new world. I don’t think being locked in the matrix by FAIs who know this is the way to keep us safe is something I reject because it is too easy. I reject it because it is essentially wireheading.
My main concern about CEV is that it winds up protecting us from nothing, that it was a waste of time and effort. But this does go along with my belief that any CEV would be one that would not incorporate my values around the importance of diversity in values, and my values around being able to reject and fight violently against other humans that had values that I found sufficiently threatening to my values.
I appreciate the questions, it is nice to sharpen my ideas a little bit. I admit I have hardly sharpened them to a mathematical precision by any means, but if you see any obvious intuition pumps working against me, I’d love to hear them.
Yes. And thank you for phrasing it that way so I understand that is at least one explanation for my concern.
It seems beyond likely to me that the CEV you get will depend heavily on just who you include in your definition of “humans” whos volition must be considered in defining CEV. Even if CEV were intended to be just that subset of volitions that “everybody” would agree on (if they were smart enough), will your definition of everybody include paranoid schizophrenics? People born with severely deformed brains? Sociopaths? Republicans? The French? My point being that our intuition is of a “common definition of human we can all agree on” but the reality of 7 billion live humans plus a few billion easy to anticipate might have a non-intuitively large variation across its volition.
So if CEV includes a “veto power” in its definition granted to all humans defined broadly enough to include sociopaths, we lose many of the values that allow us to work cooperatively.
Further concerning me, I think it is likely that humanity benefits from a diversity in values. At one level, societies with different values have different levels of success under different challenges, and in something like survival of the fittest, the societies that thrive have values that work better than those that don’t. At another level, within a society diversity in values serves the group: the nurturers are caretakers, the nerds technologists, the sociopaths become leaders and work in security.
CEV as I have heard it described sounds like a core of values, a kernel that all FAI operating systems would have to include. It doesn’t sound like a set of values or a core of meta-values that would somehow incorporate in a single kernel all the variation in values that has served humanity so well.
So yes, I am concerned that CEV is impossible, but perhaps not provably impossible, that any actual attempts to build a CEV will have more to do with the values of the people building CEV rather than some undefinable generalization of humanity.
Another concern: AI with a CEV constraint will necessarily be less adaptable than AI with no CEV constraint. So in the absence of complete totalitarian control over where AI can come from, non-CEV AI once created would eventually out-compete CEV-based AI anyway, and all that effort would have been for naught.
Finally, what I think of as a Kurzweilian paradigm of AI makes more sense to me than the idea of independent AIs that exist separately from humans. Kurzweil seems to me talks more of enhancing existing humans, building on modules, interfacing us better, and so on. Eventually, perhaps, the enhanced human is 99% enhancement and 1% human and so it becomes a matter of attitude whether you still think of it as human. Do you think CEV is something that applies to building up enhanced humans (instead of independent entities)?
The ultimate technology that makes my life too easy is wireheading. That was just fun to say, I don’t actually recognize it as a great response to your point, but I throw it out there because there might be more to it than I am allowing.
I suppose one man’s dystopia is another woman’s brave new world. I don’t think being locked in the matrix by FAIs who know this is the way to keep us safe is something I reject because it is too easy. I reject it because it is essentially wireheading.
My main concern about CEV is that it winds up protecting us from nothing, that it was a waste of time and effort. But this does go along with my belief that any CEV would be one that would not incorporate my values around the importance of diversity in values, and my values around being able to reject and fight violently against other humans that had values that I found sufficiently threatening to my values.
I appreciate the questions, it is nice to sharpen my ideas a little bit. I admit I have hardly sharpened them to a mathematical precision by any means, but if you see any obvious intuition pumps working against me, I’d love to hear them.