From: m.l.vere@durham.ac.uk
Date: Mon May 15 2006 - 05:14:34 MDT
Quoting Olie Lamb <neomorphy@gmail.com>:
> Humph. Turn your attention away from SL4 for a little while, and look what
> comes up!
>
> Let's start by clarifying some terminology. At least that will (hopefully)
> reduce the amount that people are arguing past each other.
>
> On 5/13/06, m.l.vere@durham.ac.uk <m.l.vere@durham.ac.uk> wrote:
>
> > Quoting Phillip Huggan cdnprodigy@yahoo.com:
> >
> > For the record, I define nihilist as 'one who holds no belief however
> > widespread, not supported by proof'.
>
> That's where you differ from a lot of people.
Fair play, most phrases have been used to convey several different meanings.
POI This definition was used by the russian nihilist movement as opposed to
being something i just made up to confuse people.
>
> I'd say that most ethicists (including those who identify as nihilists)
> would define nihilism as the belief that there can be no _truth_ to moral
> entities/facts/statements.
>
> That is ~~ a nihilist either "believes" or "is firmly convinced" that there
> is no objective morality.
>
> What you just described is closer to a moral sceptic - one who /doubts/
> whether there is any objective morality.
>
> What's the difference? Burden of proof. Nihilists have as much to do to
> convince a sceptic as a moral realist.
No, no 'belief', just where i see the balance of evidence to lie.
> Of course, a sceptic could also consider that some really nasty "moral
> reality" might be the case, where we "ought" to go around torturing each
> other, but if you can derive any "ought" from "is", I think you'd be hard
> pressed to show any such thing.
Yep, moral skeptic would cover where I stand as well. Ill use that term in
future to avoid the confusion of multiple definitions.
> (But then again, I'm a slightly sceptical moral realist. Anyhoo...)
>
> I'm tempted to introduce an analogy to the truth value of statements about
> the future from determinist and non-determinist. But that would be
> off-topic philosobabble :)
>
> ... Back on to SL4 relevance
> ...
>
> The reason why some people keep on barking on about Friendlyness is because
> it removes the need to quibble about meta-ethics.
>
> Considering your response to:
>
> > JohnKClark:
> >> Should? If you're right then why "should" we do anything? You are saying
> we
> >> should reject morality because it's the right thing to do and that does
> not
> >> compute.
> >
> > Should, not in terms of morality, but in terms of what will produce the
> most
> > desirable state for us (as individuals).
>
> I'd like you to consider what would, for you as an individual, produce the
> most desirable state, as far as AI motivations go...
>
> You want a superintelligent AI to behave nicely... For your future self's
> sake, you don't want it to be convinced by nihilism and do you wrong, now,
> do you?
No.
> Now, consider again your first statement: "We should reject traditional
> morality and embrace moral nihilism".
>
> If you mean "It is in our interests to (1) not assume that moral realism is
> true and (1) not assume that a clever AI won't be persuaded to be
> nihilistic", then you're dead right. It is SL4 relevant to examine whether
> a clever AI might develop a meta-ethical view that would cause it to be
> UnFriendly, and consequently shit on us.
Yep, I mean both those things. I also meant that an AI which imposes its meta-
ethics on everyone would be undesirable - as IMO there is no correct, or
single universaly most desirable set of meta-ethics.
> If you mean "Let's abandon ethics and concentrate on how we can make that AI
> do our selfish bidding," then you're a myopic fool.
see my guide AI post.
> I'll presume that you 1) are shit-stirring 2) Mean "sceptic" 3) Are actually
> considering something that impacts on Friendlyness
2 and 3 yes. 1, perhaps unintentionally.
This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:00:56 MDT