[ExI] Unfriendly AI is a mistaken idea

[Russell Wallace]

On 5/27/07, Lee Corbin <lcorbin at rawbw.com> wrote:
>
> This is just like the discussion that John Clark and I are having. I doubt
> if you'll get anyone to subscribe to the notion that they'll "will just do
> what
> we program them to".  It's trickier than that.


I think the statement is accurate given that we're dealing with one-sentence
summaries here. For example, a theorem-proving program may find proofs the
programmer didn't know about in advance - it had better, or no point in
writing it! But it's not going to say "all this math is boring, let's have
some sex and violence", not without being explicitly programmed to. In the
domain of one-sentence summaries, I think "will just do what we program them
to" is a good way to describe that.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.extropy.org/pipermail/extropy-chat/attachments/20070527/b5b4fc70/attachment.html>