Forums before death by AOL, social media and spammers... "We can't have nice things"
|    alt.cyberpunk    |    Ohh just weirdo cyber/steampunk chat    |    2,235 messages    |
[   << oldest   |   < older   |   list   |   newer >   |   newest >>   ]
|    Message 360 of 2,235    |
|    alias to joss wright    |
|    Re: AI (again)    |
|    29 Oct 03 01:21:44    |
      b6ff41ec       From: alias@removenetserver.org              On Tue, 28 Oct 2003 08:51:41 +0000, joss wright wrote:              [snip]              > i'm replying here because this ties in to my use of the "human" metaphor       > for an AI.       >       > the metaphor which i used was, i admit, based around humans. this doesn't,       > i feel, invalidate its use. the common terminology used to refer to       > intelligent agent systems uses the terms "effector" for agent routines       > which manipulate the agent's environment, and "sensor" for... well, it's       > obvious. it's easy to extend a metaphor to an AI in this way. it must       > _have_ (at least in an abstract sense) a processing center which       > correlates to the brain. similarly it must have sensors of some form (see       > below) which correlate to our own five senses.       >              i agree that for an intellegence (of any kind) to exist it must have       sensors of some kind. however i do not believe that these senses must       correlate to our own.              human senses are innapropriate for a entity existing in an environment       completely unlike our own.              (i'm cutting myself off here because A. i don't have the time to go into       it ; ) and B. u just said below that it doesn't matter all that much for       this discussion.. and i agree)                     > i think that this links in here with your point. I would argue that any       > system which is to learn _must_ have sensors. the type/style of these       > sensors is not really the issue (especially when you're being as       > conjectural with very little data as we are in this discussion/forum). if       > a system is not receiving input then it will have no new data from which       > to learn and thus the concept of intelligence is void.              agreed.. my real point here is that the senses must be relavent to the AI       if u expect it to be concerned with thier input. and IMO the sensors       generally discussed deal only with the physical world we inhabit, and as       such would be useless to a machine.                     >       > without getting into the wider issue of what intelligence/consciousness       > actually is, (what is truth, man? you heard the weirdo.), i would argue       > that any AI system which humans had a significant hand in developing would       > very likely be equipped with some form of I/O which which humans could       > interface in some fashion.              i agree that humans would be unlikely to design anything that didn't serve       us in some manner ; )              but evolution is a tricky bitch.. who sez we get to design it?              >       > if, however, you refer to an AI perceiving only a simulated universe then       > surely we would have been the creators of that universe and as such in a       > very good position to observe interact with it. we are gods in the digital       > domain.       >              actually.. i'm pretty sure we'd be fish out of water in this digital       domain. blind, deaf, and far too slow. i'm also pretty sure an AI would       have prescious little interest in a "simulated universe" .. a real       universe already exists inside that machine. that we don't find it       comfortable will likely not matter to its native inhabitants.                     > as to altering its own source code.... for a start i'd like to reiterate       > my post from another thread that a lot of current work in AI is in simpler       > systems working together rather than monolithic minds. as such, evolution       > seems perhaps a more likely approach to improving the capabilities of       > systems. for monolithic artificial mind style AI, i believe that it is       > likely that altering its own source code would probably be not of the       > greatest importance in comparison with the sorting/processing of its       > inputted data. not that i think it's an invalid point.              my point was only that *once intellegence arose* the entity would likely       have some sense of self preservation. and in a digital environment       self-preservation would include preservation of the original code that       gave birth to its consciousness.              everythings 1's and 0's right? if that code is my "mind" u can be damn       sure no ones messing with it, not even myself.              i think. dunno.. i'm not a machine ; )              -a              --- SoupGate-Win32 v1.05        * Origin: you cannot sedate... all the things you hate (1:229/2)    |
[   << oldest   |   < older   |   list   |   newer >   |   newest >>   ]
(c) 1994, bbs@darkrealms.ca