"true" AI Hardware Development
- From: "jj" <io.comp.ai@xxxxxxxxxxxxxxxxx>
- Date: Thu, 25 May 2006 01:13:34 GMT
Are you interested in developing true AI or are you just content with
simulating it on a PC?? If you want to develop true AI as opposed to
simulation of AI then you are going to have to think differently. What I
see you doing is looking in a place for something because the light is
better, instead of looking in a place where you will most likely find what
you are seeking.
We all know that the motivation for learning (the hallmark of true AI) isn't
programmed "intentionalism", it isn't programmed simulation of emotion: it
is emotional experience itself. It is emotional feelings in response to
input. Feeling Pain will motivate you into leaning what not to do. Feeling
Pleasure, on the other hand, will motivate you into learning what you can do
to repeat the emotional sensation. These are two of the primal learning
forces for intelligence possessed by living organisms. We need to emulate
life's hardware to process this analog data efficiently.
Engrams are stamped with particular emotion(s) and are likewise encoded with
the input information. Memory retrieval works in much the same way and
associative recall and retrieval from memory can index memory by emotion.
Recall all the happiest times lately that you have had. Recall the most
painful. They can be retrieved by remembering the states of emotion they
included when those thoughts were recorded.
Emotions are analog in nature. Computers process things using binary logic.
There is an obvious mismatch in hardware/software/data. We can create
conversion software to input/output the data, but it would have such severe
limitations that the best you could hope for is an incredibly retarded
computer. Also, the analog data would be simply objective data instead of
subjective or analog driven. Clearly, what is needed is a completely
compatible machine capable of being driven by analog data. Am I the only
one that sees this?
I have come a long, long way to developing such a machine. It will be an
analog processor with capabilities to input analog information, process and
output analog information. It will possess experiential capabilities to
drive the learning process. It will be capable of becoming self-aware. The
state of being self-aware, incidentally is not the final step of Artificial
Intelligence; it is the first step. Programming it is not like programming
a computer. Once the machine achieves self-awareness, it starts to program
itself, if it possesses the compatible memory and peripheral input and
output devices (all analog, of course).
Where can I get funding for such a machine?? I have exhausted all of my
time and money these last few years and I am not getting any younger. If
any of you think this is a worthwhile project and know who would be
interested in funding it, I can produce designs and apparatus to show that
it is a real ongoing project capable of coming to fruition quickly. I have
already developed a good base and have developed solid hardware towards this
goal. So, please let me know. I'm just inquiring.
Thanks for reading this post.
i/o
.
- Follow-Ups:
- Re: "true" AI Hardware Development
- From: jalegris
- Re: "true" AI Hardware Development
- From: David Formosa (aka ? the Platypus)
- Re: "true" AI Hardware Development
- From: Joachim Pimiskern
- Re: "true" AI Hardware Development
- From: Yeoman
- Re: "true" AI Hardware Development
- Prev by Date: Re: Goal of AI: Perfect or Bounded Rationality
- Next by Date: Re: Goal of AI: Perfect or Bounded Rationality
- Previous by thread: Goal of AI: Perfect or Bounded Rationality
- Next by thread: Re: "true" AI Hardware Development
- Index(es):
Relevant Pages
|