From: Eliezer S. Yudkowsky (sentience@pobox.com)
Date: Wed Feb 06 2002 - 01:17:28 MST
Gordon Worley wrote:
>
> On Wednesday, February 6, 2002, at 01:43 AM, Mitch Howe wrote:
>
> > What I'm wondering is if anyone has any ideas about how a Sysop might
> > be supremely well protected from hacking. My background isn't
> > technical enough to give any real-life examples besides "physical
> > separation" Could something as upiquitous as a Sysop realistically do
> > this?
>
> First off, it would be a major violation of volition to hack the Sysop.
> The Sysop will most likely not want to be hacked, therefore ve cannot be
> hacked. But, that doesn't mean an attack couldn't still be tried.
Mm, that sounds like circular logic to me. The Sysop is what supposedly
enforces nonviolation of volition. So you can't invoke "violation of
volition" rules to explain why the Sysop itself is protected against a
perversion attack. To discuss a perversion attack against the Sysop, you
need to revert from the high-level rules governing the content of the
Sysop Scenario to the low-level rules governing the implementation of the
Sysop Scenario.
-- -- -- -- --
Eliezer S. Yudkowsky http://intelligence.org/
Research Fellow, Singularity Institute for Artificial Intelligence
This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:00:37 MDT