4 ms·
How do you know one would eventually?
by perfmode 7y ago
How do you know one would eventually?
- UnFleshedOne 7y agoGood question. There is no reason to think otherwise at the moment and research in physics, biology, neuroscience, psychology and sociology kinda points that way? And I know that as well as I know anything I'm not an expert in -- from my own cursory review and stated opinions of people who are experts (again, as far as I know). In other words, it would be surprising otherwise? Our morals/desires depend on the way we are built and they dictate the 'oughts', thus providing a link from 'is'. And of course it is possible that we can never have enough information to decompile our own values because it is lost in compression, so we'll be stuck with dealing with a black box. But this is a technical limitation, not a philosophical one, and for a simpler system (for example a AGI if one can be built understandably) we would be able to derive its 'oughts' from what it 'is'.
- perfmode 7y agoYour prediction without proof is about as bankable as a prediction without proof of P == NP or P != NP. Just because something is intuitive doesn’t make it true.