RE: What kinds of things?
- Bob Haugen asks:
> From: Panu ViljamaaBob:
> > Going in the direction from "How to build?" to "What to
> build?", we should
> arrive at the question "Why to build?"
> In the midst of the last few messages, I started to wonder,
> what do people have in mind in this discussion?
> * autonomic systems (self-healing etc)
> * autonomous intelligent agents
> * intelligent assistants to humans
> * weak "artificial life"
> * evolving software
> * "strong AI"
> * intelligent replacements for humans
> * something else?
> Not that the categories above are necessarily
> mutually exclusive...
Good question. In my view, the answer is all of the above
but probably less:
- there is a wide range of diversity: some agents
are more intelligent than others but none reach
the "strong AI" level
- despite this diversity there are much larger
populations of "dumb" agents that operate on
very simple rules i.e. like neurons
- perceived overall intelligence is achieved by the
interaction of the mostly "dumb" agents
- some agents operate unaffected by humans
i.e. like internal healing systems
- some other agents do interact with humans
but they do not attempt to replace human
All of this still within the spirit of a "Living Metaphor":
lots of dumb things interacting through simple rules.