[ExI] Robust and beneficial AI

John Clark johnkclark at gmail.com
Mon Jan 19 17:57:32 UTC 2015


On Mon, Jan 19, 2015 William Flynn Wallace <foozler83 at gmail.com> wrote:

​> It doesn't matter if I understand it or if I implement it or not, if
>> Mr.Jupiter Brain wants it done it will get done.
>>
>
> > What about my other point?  When the AI is wrong?
>

If the error is so subtle that even the super intelligent AI can't catch it
then you're not going to catch it either. And I'm going to go out on a limb
here and make a prediction, mistakes will be made in the future.

 John K Clark
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.extropy.org/pipermail/extropy-chat/attachments/20150119/a0cc53e0/attachment.html>


More information about the extropy-chat mailing list