From: John Donaldson <John.Donaldson.nul> Date: Wed, 2 Jan 2013 19:35:02 +0000 Archived: Thu, 03 Jan 2013 07:06:13 -0500 Subject: Re: Artificial Intelligence >From: Ray Dickenson <r.dickenson.nul> >To: <post.nul> >Date: Mon, 31 Dec 2012 20:55:03 -0000 >Subject: Re: Artificial Intelligence >Hello List >Have to start over because it seems we're forgetting a couple of >things: >First, 'morals/ethics' or whatever only exist because humans >(and maybe other higher mammals) have taboos which have >developed for evolutionary reasons. That is, _all_ of those >"sins" tend to decrease the gene pool, even the one of usury >(because usurers aren't working or competing they tend to become >ugly and incompetent, eventually unmarriageable and the line >goes extinct). I.e. "sins" decrease our species chances of >survival, especially in times of genetic bottle-neck. >So if an AI is actually 'intelligent' and not merely following a >program (algorithm) it will have its own definition of morals, >based on its own evolutionary imperatives - which are almost >certainly going to clash with ours. >Second, as Penrose has pretty firmly concluded (in Emperors New >Mind & Road to Reality) conscious intelligence looks likely to >depend on a non-algorithmic and indeed non-computable ability >which _might_ only belong to organic brains. [He speculates it >might be a quantum attribute.] >So the successful AI might have to have an organic (neural >network) brain - which brings us back to its evolutionary >imperatives and hence its - unknown to us - morals and motives. There's certainly an interesting debate to be had about realism vs. anti-realism in ethics, although it's not the sort of debate that can be settled easily - see, e.g: http://plato.stanford.edu/entries/moral-realism/ Given the complexity of the debate, I would urge epistemic caution in making sweeping, far-reaching claims about the nature of ethics. For instance, the argument presented above is rather fast and loose. It might be true that evolutionary forces have shaped human psychology such that ethical behaviour is brought about, but that doesn't support anti-realism about ethics. Evolutionary forces have shaped the visual system, but that doesn't mean light isn't real. Moreover, it is notoriously difficult to establish an evolutionary hypothesis for the genesis of this or that specific trait - "just-so" stories are easy to tell, but they should be treated with caution - see: Gould and Lewontin's classic paper .pdf: http://tinyurl.com/axfyu7k Similarly, I would urge epistemic caution in coming to sweeping conclusions when considering the question: "is the mind a computer?" This is a difficult to question to make clear, never mind settle - see: http://plato.stanford.edu/entries/computational-mind/ Cognitive science is in its infancy, if that. Best wishes, John Listen to 'Strange Days... Indeed' - The PodCast At: http://www.virtuallystrange.net/ufo/sdi/program/ These contents above are copyright of the author and UFO UpDates - Toronto. They may not be reproduced without the express permission of both parties and are intended for educational use only.
[ Next Message | This Day's Messages ]
This Month's Index |
UFO UpDates - Toronto - Operated by Errol Bruce-Knapp