From: Ben Goertzel (ben@goertzel.org)
Date: Wed Jan 26 2005 - 06:18:49 MST
> > I wonder if any of you
> > would care to agree/disagree/comment on the
> > following:
> >
> > 1. In developing FAI, some form of Last
> > Judge/AI Jail will be used.
It's a possible approach, but not necessarily the best way...
> > 3. It would be intractible to audit the "grown"
> > code of a seed AI. That is, once
> > the AI begins to self improve, we have to
> > evaluate its (I refuse to use
> > posthuman pronouns) degree of friendliness on
> > the basis of its actions.
Not entirely true. It is already intractable to audit the code of large
software projects, though.
If the seed AI is created right, then evaluating it will be a combination of
evaluating its behaviors with evaluating its internal structures.
-- Ben G
This archive was generated by hypermail 2.1.5 : Tue Feb 21 2006 - 04:22:52 MST