[Synapse elist] my first attempt at coming up with some answers

[Daniel Bisig]

Hi all,

since my previous email was an attempt to explain some of my own  
motivations and activities in the field, it totally missed out  
concerning the interesting questions that have been posed by Reva. In  
this second email of mine, I will try to give a go at these questions.

What role does science and technology play in designing life?

It is interesting to note how tightly integrated science and  
technology has become in any endeavor to design life. While this  
combination seems essential in order to develop both understanding and  
skills in creating or modifying life, it results at the same time in  
an uneasy proximity to a profit and commodity attitude. Concerning the  
life sciences, this attitude seems to create a pressure towards a  
constriction and limitation of biological life's natural capabilities.  
The synthetic sciences of AI / ALife on the other hand press forward  
to include more and more of lifelike capabilities in artificial  
systems. It looks like there might come a point were the two  
tendencies cross and modified biological life is more impaired than  
entirely artificial organisms.

Are biologically based intelligence and silicon based artificial  
intelligence systems structurally similar?

I definitively think so. I believe that intelligence is based on a  
fundamental repertoire of structural and behavioral components that  
are capable of establishing and maintaining feedback systems between  
bodily and mental states on one hand and external situations on the  
other. And as long as the biological and artificial intelligences act  
in very similar environments, a high level of similarity is imposed by  
the simple fact that both intelligences have to perceive and react to  
the same ultimately physical cues. So as long as artificial  
intelligences exist as embodied entities in a real environment, this  
similarity propagates from the agent's outer self towards its inner  
processes, becoming diluted in the process but never disappearing  
entirely.

Can the informational aspects of consciousness be extracted from the  
material?

To me this questions sounds very much like the physical symbol systems  
hypothesis but rephrased by using the term consciousness. This  
impression might be totally wrong but I try to replay to this question  
based on this interpretation. It is the assumption, that it is mainly  
informational aspects that make up intelligence, that lies at the root  
of classical AI and that has been questioned and criticized by many.  
After all, embodied AI represents one of the alternative approaches  
that has emerged based on the assumption that material and information  
properties are tightly integrated and interdependent. The same should  
hold true concerning the relationship between consciousness and  
material. But this leads to another question that might be interesting  
to discuss: what is the relationship between intelligence and  
consciousness. Does consciousness necessarily depend on a high level  
of intelligence and does it naturally emerge once that level of  
intelligence is achieved? How can AI contribute to this question? Can  
we assume at all that AI can complement cognitive sciences and  
neurosciences when it comes to this question? If the property of  
consciousness is always assumed purely based on interpretation of an  
observer, we immediately bump into the frame of reference problem.

Is the physical world entirely rule based and therefore duplicatable?

As long as one assumes that the physical world is subject to natural  
laws and that one knows all the natural laws at a particular level of  
organization and that one is capable of micking these laws in some  
artificial system or alternate universe, then one has succeeded in  
duplicating the physical world. But this is a duplication on a  
functional level, it is not a duplication of a state of the world, let  
alone a duplication of the future change of the world. Chaos theory  
and uncertainty principles obviously oppose such a possibility.

Can computer-based codes for interactivity cross over into living  
agency?

I must admit that I'm fairly lost at that question. What do you mean  
by cross over in this context? I would very much like to try to  
brainstorm about this question since it seems very important. The  
capability for interaction lies at the very hearth of being alive. And  
it is this capability, the represents the most important strategy to  
create at least an illusion of a living system. For this reason,  
interactivity can be very deceptive and confusing. It is therefore at  
the level of interactivity, were an artist, scientist or engineer  
reveals his/her sincerity in creating a living agency.


Best regards

      Daniel

-------------------------------------------

Dr. Daniel Bisig

Artificial Intelligence Laboratory
University of Zurich
Andreasstr. 15
CH-8050 Zurich
Switzerland
T: +41 44 635 45 77
F: +41 44 635 45 07
E: dbisig at ifi.uzh.ch
W: bitingbit.org

Institute for Computer Music and Sound Technology
University of the Arts Zurich
Baslerstrasse 30
CH-8048 Zürich
Switzerland
T: +41 43 446 55 18
F: +41 44 578 78 10
E: daniel.bisig at zhdk.ch
W: bitingbit.org

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.synapse.net.au/pipermail/elist/attachments/20080605/c2eaae39/attachment.html