Re: Forbidden Planet

From: Keith Henson (
Date: Tue May 11 2004 - 23:44:07 MDT

A list reader wrote:


> As for the stanford experiment, what do we know about those that took
> part, were they volunteers? Did the guards volunteer to be guards?

Excuse me for being annoyed at people who write before clicking on the
link. The answer to the question is two clicks in from here:

A Simulation Study of the
Psychology of Imprisonment
Conducted at Stanford University

Welcome to the Stanford Prison Experiment web site, which features an
extensive slide show and information about this classic psychology
experiment. What happens when you put good people in an evil place? Does
humanity win over evil, or does evil triumph? These are some of the
questions we posed in this dramatic simulation of prison life conducted in
the summer of 1971 at Stanford University.

How we went about testing these questions and what we found may astound
you. Our planned two-week investigation into the psychology of prison life
had to be ended prematurely after only six days because of what the
situation was doing to the college students who participated. In only a few
days, our guards became sadistic and our prisoners became depressed and
showed signs of extreme stress. Please join me on a slide tour of
describing this experiment and uncovering what it tells us about the nature
of Human Nature.


What suspects had done was to answer a local newspaper ad calling for
volunteers in a study of the psychological effects of prison life. We
wanted to see what the psychological effects were of becoming a prisoner or
prison guard. To do this, we decided to set up a simulated a prison and
then carefully note the effects of this institution on the behavior of all
those within its walls.

More than 70 applicants answered our ad and were given diagnostic
interviews and personality tests to eliminate candidates with psychological
problems, medical disabilities, or a history of crime or drug abuse.
Ultimately, we were left with a sample of 24 college students from the U.S.
and Canada who happened to be in the Stanford area and wanted to earn
$15/day by participating in a study. On all dimensions that we were able to
test or observe, they reacted normally.

Our study of prison life began, then, with an average group of healthy,
intelligent, middle-class males. These boys were arbitrarily divided into
two groups by a flip of the coin. Half were randomly assigned to be guards,
the other to be prisoners. It is important to remember that at the
beginning of our experiment there were no differences between boys assigned
to be a prisoner and boys assigned to be a guard.

(end web site quotes)

>As for how this applies to any future AI, I think we have to keep in mind
>that we're not necessarily trying to re-create the organisation of the
>human brain in a piece of software, it is just that the human brain and
>the human condition are all we have as a model of intelligence - luckily
>there's enough intelligence in some of these brains that I think we can
>avoid our shortcomings when creating an AI,

Assuming I am right about the evolutionary origin of this trait in humans,
it was figured out less than two weeks ago. My point, in case you missed
it, was that we *don't know* how many other hideously dangerous
psychological quirks we have *or* what turns them on. Miss one and
accidentally turn it on in a powerful upload and you could get irrational
powerful beings---

I think I would rather take up of smoking in a powder magazine.

Keith Henson

>one of the reasons I think AI is so fascinating.

This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:00:46 MDT