[ExI] Digital Consciousness .

Anders Sandberg anders at aleph.se
Wed Apr 24 21:05:52 UTC 2013

On 24/04/2013 17:06, Giovanni Santostasi wrote:
> Have you seen this recent paper that show "intentionality" can 
> actually be realized by a simple law that looks similar to a 
> thermodynamical entropic force?
> http://physics.aps.org/articles/v6/46

This is not the same kind of intentionality we have been discussing in 
this thread. This is closer to having intent.

Acting so you get maximal future options seems to be a good strategy in 
general, although in specific cases it gets tricky. Think of Ulysseus pacts,
that however are tools for ensuring more long-term options or value. 
Chess is another example where you actually do not want to go for more 
open futures - you want to get to a future where your opponent loses. 
Aiming for maximally open futures makes sense if you want to play the 
game indefinitely.

Intent-like behavior can be generated from simple rules (think of 
evolution), but does not have to have goals. In the paper goals were 
inherent in the setup. It can also lack intentionality (having 
representations), for example as in the case of evolution. When we have 
intent we typically have representations and goals, where the goals are 
usually chosen by us via earlier intentional mental actions. Of course, 
whether there is any real difference between the cases or just a long 
continuum is at the heart of an endless debate.

Anders Sandberg,
Future of Humanity Institute
Philosophy Faculty of Oxford University

More information about the extropy-chat mailing list