Re: AI Goals [WAS Re: The Singularity vs. the Wall]

From: Jef Allbright (
Date: Tue Apr 25 2006 - 18:10:57 MDT

On 4/25/06, BillK <> wrote:
> On 4/25/06, Jef Allbright wrote:
> > Effectively promoting shared human values into the future requires
> > effectiveness in two complementary areas:
> >
> <snip ivory-tower idealism>
> >
> > Increasing awareness of what works, applied to increasing awareness of
> > our shared human values, leads to increasingly effective
> > decision-making that will be seen as increasingly good.
> >
> > This broad-based framework based on human values, amplified by
> > technology (including AI) will effectively promote our values into the
> > future.
> >
> 'shared human values' don't exist. This is just wishful thinking on your part.

Bill, it seems that you snipped my supporting statements. I would
very much appreciate useful feedback on this thinking. Here are some
examples of commonly shared (not necessarily universal) values.

Low level shared values:
 We value the safety of our children.
 We value the growth of our children.
 We value social status.
 We value respect given to us.
 We value good health.
 We value strength.
 We value insiders over outsiders.
 We value sexual enjoyment.
 We value altruism.
 We value reciprocity.
etc., etc.

Cultural level shared values:
 We value freedom from physical harm.
 We value marital fidelity.
 We value personal property.
 We value respect for proper authority.
 We value fairness (equity.)
 We value sexual attraction.

Higher level shared values:
 We value truth.
 We value rationality.
 We value consistency.
 We value coherence.
 We value good over evil (as we see it.)

> If the first AI is developed in China (or worse still, Iran) you will
> quickly find out whose human values you will be sharing.

Exactly why increasing positive-sum interactions is increasingly important.

> And if the first AI is developed by the US Military then half the
> world will be opposed to the culture it will be supporting.

And why AI and IA, in the service of shared human values over
increasing scope will be seen as increasing the good.

And as you deleted it from my previous post while trying to show that
I am overly idealistic, I will repeat that these efforts will be in
vain if we destroy ourselves first. We are at significant risk from
biotech, nanotech, and AI tech, currently lacking the wisdom to
develop, deploy and counteract these technologies safely.

- Jef

This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:00:56 MDT