From: Phil Goetz (firstname.lastname@example.org)
Date: Mon Aug 22 2005 - 10:06:58 MDT
> ...is mentioned prominently on the SL4 list information
> page that you must have visited when you signed up.
But not on www.sl4.org .
> Hopefully you will
> locate the previous discussion of why posting is (a)
> and (b) would defeat the point of the exercise.
I located a posting with Eliezer's reason for keeping
the results secret. Which he of course can do. And while
my estimated prior for changing Eliezer's mind on this
subject is very low, I'd like to give these reasons why he
might want to:
1a. I am familiar with many independent and
well-substantiated cases of appearances of the Virgin Mary.
I give them small probabilities of truth, because I gave
them small priors. I'd need details of each case to
overcome these low priors. Likewise with the AI-box.
1b. Someone once said that a scientist is someone who,
instead of believing something because he sees it, is
someone who believes something because he has a theory for
it. I can't construct a theory for the AI-box results
without more details of what happened.
2. I want to know whether Eliezer is using the same
"trick" each time.
3. The outcome is meant to suggest that, if Eliezer can
convince an arbitrary human to do something they are
dead-set against doing, then so can an AI. But if Eliezer
had such powers, why can't he consistently convince people
on SL4 to change their views?
This suggests a solution to boxing an AI: Create a list
advocating the position that AIs will be friendly. Wait
until that list's Marc Geddes-equivalent appears. Have him
be the gatekeeper.
4. I'm interested in the transcripts for what they say
about human psychology.
5. Since there have been 3 experiments already,
perhaps one of them could be described in detail.
At least one other participant would have to change his
mind in order to release the results. But I'm sure Eliezer
could talk him into that. :)
- Phil Goetz
Start your day with Yahoo! - make it your home page
This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:00:52 MDT