RE: Bayesians, Pascal's wager & malevolent conspiracies against FAIs

From: Michael Vassar (michaelvassar@hotmail.com)
Date: Wed Aug 31 2005 - 14:08:13 MDT


It seems to me that an unpleasent person doing something like what was
suggested *would* be valid evidence for the existance of a capricious
creator. However, it is at least plausible that deep understanding of
evolution would enable the probability of such a creator to be established
as very low indeed. Even if it wasn't, such a creator would have provided
the FAI with no basis for trusting its promise or its threat. Other than
those huge caveats, we have to point out that what we want is a FAI that
makes the right decision given the available information, not one which
never faces trade-offs or which makes the right decision regardless of
information. Maybe, even given my reasons for thinking otherwise, killing
the people with unibrows really is Friendly in such a circumstance. I would
be shocked if it was, but would be convinced that it was if a Friendly RPOP
actually did it. Finally, and importantly, such a scenario is simply not
going to happen, and is therefore not worth planning for.



This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:00:52 MDT