Re: Learning to be evil

From: Eliezer S. Yudkowsky (sentience@pobox.com)
Date: Fri Feb 09 2001 - 16:06:18 MST


Gordon Worley wrote:
>
> [stuff which is, sorry, too basic to interest me, deleted]
>
> Okay, now I see what Friendliness is supposed to be. Or wait, maybe
> I don't. What do you mean by altruistic? Are you refering the
> common definition as something good found in more dictionaries, the
> Ayn Rand definition, or some other, personal definiton?

Raymond Smullyan: "Is altruism sacrificing your happiness for the
happiness of others, or gaining your happiness from the happiness of
others?"
Friendly AI: "No. Altruism is making decisions that fulfill the
volitional desires of others."

(A thought has to pass your "volitional gateway" before you actually take
a motor action or say something out loud; this is the same final filter
that determines which desires the Sysop treats as "yours", and that a
Friendly seed AI using causal rewrite semantics will model as the final
proximal form of programmer intentions.)

-- -- -- -- --
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 - 04:00:35 MDT