Ai Forums Home Welcome Guest    Friday, August 18, 2017
Ai Site > Ai Forums > Language Mind and Consciousness > Interesting that matter has evolved to ... Last PostsLoginRegisterWhy Register
Topic: Interesting that matter has evolved to contemplate itself

3decimal1415
posted 2/6/2012  00:48Send e-mail to userReply with quote
Interesting that matter has evolved to contemplate itself. When will we know that we have created AI systems are conscience of themselves?


3decimal1415
posted 2/6/2012  01:01Send e-mail to userReply with quote
 
3decimal1415 wrote @ 2/6/2012 12:48:00 AM:
Interesting that matter has evolved to contemplate itself. When will we know that we have created AI systems are conscious of themselves?

 
sorry, I meant conscious not conscience...


tkorrovi
posted 2/6/2012  18:45Send e-mail to userReply with quote
 
3decimal1415 wrote @ 2/6/2012 12:48:00 AM:
When will we know that we have created AI systems are conscience of themselves?

 
We can never know that because one aspect of consciousness is subjective experience which we cannot get any information about and thus also cannot model it.

However, we can know whether a system has certain aspects of consciousness like these defined by Bernard Baars, because these are objective and defined and thus their presence can be tested.

When the known objective aspects of consciousness are present, we can say that the system is an Artificial Consciousness. A system where all these aspects are present is all which we can achieve. We may assume that such system also has a subjective experience but we cannot test it.

 Artificial Consciousness ADS-AC project
Last edited by tkorrovi @ 2/6/2012 6:55:00 PM

simnia
posted 9/2/2012  18:28Send e-mail to userReply with quote
 
3decimal1415 wrote @ 2/6/2012 12:48:00 AM: [spelling corrected]
When will we know that we have created AI systems are conscious of themselves?

 
Normally I don't like debating things that are vaguely defined like "consciousness", but since you said "conscious of themselves", you're talking about "self-awareness", which is a much more clear-cut concept.

However, there is one more major definitional hurdle here that needs to be cleared up: Exactly which part of itself would you mean by "aware of itself"? I see two parts that are significant that you might be intending: (1) body; (2) processor (mind).

(1) Body. If you mean aware of the physical presence and location of the body of an autonomous vehicle or robot, that is almost trivial, since such a robot mapping out its surroundings will probably need to know where its own body is, which is almost trivial, even if there is merely a blank spot on the map at that location. For example, the diagrams at...

http://www.uq.edu.au/think/robots
http://www.societyofrobots.com/programming_wavefront.shtml
http://www.ribbonfarm.com/2007/07/20/harry-potter-and-the-cuaron-slam/

...are the kind of diagrams I mean, which show both the robot and its environment, although offhand I can't find an online reference to such a map created by the robot itself.

(2) Processor. If you mean aware of the processor and/or memory of the robot, that is only slightly more difficult: you just put a sensor into a circuit or memory to monitor what is going on there, like that a circuit is functioning with electricity, or that data exist in a given range of memory cells. That may not be a useful thing to do, but it could be done with no theoretical difficultly at all. Self-diagnostic tests probably already do that.

Either way, it is easy to make a robot "self aware". So in answer to your question, the way we will know is that we will have designed it with those features and the presence of those features will be obvious. If those features I mentioned aren't what you consider implementation of "self-awareness" (which I don't consider them to be, at least not in the way that I think you mean it), then you should reconsider your definitions and rephrase your question so that it gets to the heart of the matter. Another possibility is that the issue of self-awareness isn't even important.


Last edited by simnia @ 9/3/2012 6:41:00 PM
  1  
'Send Send email to user    Reply with quote Reply with quote    Edit message Edit message

Forums Home    The Artificial Intelligence Forum    Hal and other child machines    Alan and other chatbots  
Contact Us Terms of Use