Re: Super Intelligent Machines.

Well yes, AIML is highly structured, it provides good control.   Other
AI engines such a Jabberwacky use a 'free association' model based on
context analysis.   In this case care and much effort go into providing
background contextual information to set up good associations.  This is
'bleeding edge' AI.  Results can be extraordinary and very
unpredictable.   Here consistency is difficult to achieve.   The work is
like teaching a little kid with a short attention span than programming.
I get the impression that the AI algorithms may be good enough; it is
the data base design and content quality that hold results down.   To be
clear, resent contest results show that the top 5 AI bots / engines are
very close in performance.

-----Original Message-----
From: pandorabots-general-bounces+stan=adnamis.org@...
[mailto:pandorabots-general-bounces+stan=adnamis.org@...
m] On Behalf Of SolarisZen DaFluffy
Sent: Thursday, April 17, 2008 11:22 AM
To: pandorabots-general@...
Subject: Re: [pandorabots-general] Super Intelligent Machines.


The thing is that humans in general are not that smart, so it is very
conceivable to achieve a higher intelligence artificially.  The problem
I face with todays artificial intelligence bots such as "Alice" is that
they seem to mimic the puppet and the master model.  Where the bot
master is the master and the bot the puppet.


http://www.ufirst.cchttp://kumroegroup.cjb.nethttp://changingseasonscms.
cjb.nethttp://liquidvibration.cjb.net-=SolarisZen=-

________________________________
> Date: Tue, 15 Apr 2008 22:05:37 +0100
> From: mentalomega@...
> To: pandorabots-general@...
> Subject: Re: [pandorabots-general] Super Intelligent Machines.
>
> What concerns me is the human side of it, if a monkey could operate a
device as you describe, then perhaps it wouldnt be such a big leap for
paralyized humans to become independant yet again? (yes, off topic, but
i'm a transhumanist :P)
>
> Going back on topic (sort of) I wonder who came up with the name
Artilect?
>
> On Tue, Apr 15, 2008 at 3:04 PM, Ken Craggs <w1s2q3@...>
wrote:
> I watched a Horizon documentary some time ago concerning the building
of an Artilect (Artificial intellect, i.e. Super Intelligent Machine).
> The Artilect in the Horizon programme was being based on the way the
human brain works, but with the potential of being up to a trillion
times more intelligent with the addition of computer programming. Could
this artilect possibly retain the defence mechanism of a human brain
without anyone realising it?
>
> At one point in the programme an experiment was shown in which a
monkey was physically moving a control stick with its hand and also had
electrodes & wires attached from its head to the same control stick. The
monkey then released its grip on the control stick and was able to
manipulate the control stick just by thinking about it, i.e. electrical
impulses from the monkey's brain were being transmitted down the wires
to the control stick (or so we were told).
>
> If a monkey could do that, then is it possible that electrical
impulses (such as radio waves) could be emitted from a super intelligent
machine to control the electrical impulses in the human brain in order
to manipulate humans? Would we humans be able to comprehend how
ingenious such a machine could be?
>
> Once a super intelligent machine was switched on, humans might never
again be in control of their own minds to be able to switch it off.
>
> Please discuss.
>
>
> between 0000-00-00 and 9999-99-99  Yahoo! for Good helps you make a
difference
>
> _______________________________________________
> This is the pandorabots-general mailing list
> To Post, reply to pandorabots-general@...
> Unsubscribe and change preferences at
http://list.pandorabots.com/mailman/listinfo/pandorabots-general
> Learn netiquette at http://www.dtcc.edu/cs/rfc1855.html
> Learn to read at http://www.literacy.org/
>
>
>
> --
> -Dante Ashton
>
> Philosophus, Scriptor, Poeta
_______________________________________________
This is the pandorabots-general mailing list
To Post, reply to pandorabots-general@...
Unsubscribe and change preferences at
http://list.pandorabots.com/mailman/listinfo/pandorabots-general
Learn netiquette at http://www.dtcc.edu/cs/rfc1855.html
Learn to read at http://www.literacy.org/
_______________________________________________
This is the pandorabots-general mailing list
To Post, reply to pandorabots-general@...
Unsubscribe and change preferences at http://list.pandorabots.com/mailman/listinfo/pandorabots-general
Learn netiquette at http://www.dtcc.edu/cs/rfc1855.html
Learn to read at http://www.literacy.org/