On 2015-10-14 15:39, EdFromNH . wrote:
Re uploading, I should think those who are seeking to make "safe" AI
should be rather afraid of uploaded minds, because of their potential
to merge into machine superintelligence, while presumably maintaining
all the primal, often conflicting, and largely selfish instincts of
human nature.
Bostrom argues in his book on the topic that the technology required to
implement
uploads would probably lead instead to what he calls 'neuromorphic AI' -
which I
think refers to neural networks. He identifies these as being relatively
difficult to
control and as lacking in transparency.
I think the correct conclusion is not that we should be afraid of
uploads - but
rather that we should forget about uploads of individual minds playing a
significant role. By the time uploads become practical, other forms of
machine intelligence are likely to have already surged ahead, and the
uploads
will be obsolete.
--
__________
|im Tyler http://timtyler.org/
-------------------------------------------
AGI
Archives: https://www.listbox.com/member/archive/303/=now
RSS Feed: https://www.listbox.com/member/archive/rss/303/21088071-f452e424
Modify Your Subscription:
https://www.listbox.com/member/?member_id=21088071&id_secret=21088071-58d57657
Powered by Listbox: http://www.listbox.com
