An agent optimized to humanity's CEV would instantly recognize that trying to skip ahead would be incredibly harmful to our present psychology; without dreams—however irrational—we don't tend to develop well in terms of CEV. If all of our values break down over time, a superintelligent agent optimized for our CEV will plan for the day our dreams are broken, and may be able to give us a helping hand and a pat on the back to let us know that there are still reasons to live.

This sounds like the same manner of fallacy associated with determinism and the ignorance of the future being derived from the past though the present rather than by a timeless external "Determinator."

I think you're vastly underestimating the magnitude of that "helping hand."

By way of analogy... a superintelligent agent optimized for (or, more to the point, optimizing for) solar system colonization might well conclude that establishing human colonies on Mars is incredibly harmful to our present physiology, since without oxygen we don't tend to develop well in terms of breathing. It might then develop techniques to alter our lungs, or alter the environment of Mars in such a way that our lungs can function better there (e.g., oxygenate it).

An ag... (read more)

What makes us think _any_ of our terminal values aren't based on a misunderstanding of reality?

by bokov 1 min read25th Sep 201389 comments


Let's say Bob's terminal value is to travel back in time and ride a dinosaur.

It is instrumentally rational for Bob to study physics so he can learn how to build a time machine. As he learns more physics, Bob realizes that his terminal value is not only utterly impossible but meaningless. By definition, someone in Bob's past riding a dinosaur is not a future evolution of the present Bob.

There are a number of ways to create the subjective experience of having gone into the past and ridden a dinosaur. But to Bob, it's not the same because he wanted both the subjective experience and the knowledge that it corresponded to objective fact. Without the latter, he might as well have just watched a movie or played a video game.

So if we took the original, innocent-of-physics Bob and somehow calculated his coherent extrapolated volition, we would end up with a Bob who has given up on time travel. The original Bob would not want to be this Bob.

But, how do we know that _anything_ we value won't similarly dissolve under sufficiently thorough deconstruction? Let's suppose for a minute that all "human values" are dangling units; that everything we want is as possible and makes as much sense as wanting to hear the sound of blue or taste the flavor of a prime number. What is the rational course of action in such a situation?

PS: If your response resembles "keep attempting to XXX anyway", please explain what privileges XXX over any number of other alternatives other than your current preference. Are you using some kind of pre-commitment strategy to a subset of your current goals? Do you now wish you had used the same strategy to precommit to goals you had when you were a toddler?