From: Christopher Healey (CHealey@unicom-inc.com)
Date: Mon May 02 2005 - 09:19:48 MDT
You're right, we face some high hurdles, but I'm not sure the answer is to stop trying to clear them; adopting the belief that we will successfully end up at the other end of the track. We're just not that good, and reality *doesn't* care.
Question 1: If Friendliness is likely to arise anyway, what are the consequences of pursuing it with due speed?
Question 2: If Friendliness is not likely to arise, what are the relative consequences of not pursuing it?
FAI is an exercise in risk management, and no matter what we can or cannot agree on, our actions must adress the *actual* outcomes we seek to achieve.
When faced with such questions, I now try to always ask myself: "Does this view I hold actually make any useful prescriptions toward achieving my goals?", and if it doesn't, "Why do I hold a view that offers no guidance on acting to achieve my goals?"
From: email@example.com on behalf of Tennessee Leeuwenburg
Sent: Mon 5/2/2005 8:12 AM
Subject: Beating the rush
-----BEGIN PGP SIGNED MESSAGE-----
History is fun to watch. Every week I see some new development
pointing towards a technological step-improvement in the state of
technology. Many of these changes could be made use of in
Question : Do we have time to solve Friendliness before we are
confronted with the first real AIs?
My answer : No.
Do people here agree, and what is to be done?
My own position is that morality, if not friendliness, is likely to
arise anyway. But many disagree, and for those who do, such a
judgement may seem as hollow as faith.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.0 (MingW32)
Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org
-----END PGP SIGNATURE-----
This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:00:51 MDT