I was attempting the problem of classifying whether a bar code scanner is example of artificial intelligence or not. I have become very confused after thinking on this:
I am under the impression that bar code scanner doesn't seem to be intelligent, as it just does what it is programmed to do. Agreed that it takes percept, processes and peforms an action, but then why my program, accepting a number as input from the user and outputting whether it is odd or even, cannot be considered intelligent?
But I am not even able to convince myself that it is not intelligent. Because I am under the impression that the search algorithms, the vacuum cleaner and all which we consider as intelligent are nothing just a clever programming.
Even the more complex activities like facial recognition which we consider as AI, is nothing but a bunch of neural networks just learning the right weights by using mathematics. Or the cancer prediction is just drawing a line drawn based on previous data. I don't know, but it doesn't feel intelligent to me. It appears that complex appearing tasks are called Ai, while simpler automation is not.
This again leads me to think what I might consider as intelligent? Consciousness is the first word that comes to my mind, but again I find myself incapable to exactly define what consciousness is? Even taking the inherent understanding of it, leads me to certain contradictions. I have been told that plants are conscious but it doesn't appear to me that they display any sort of consciousness. Like a sunflower cannot "consciously ' decide not to turn towards sun, while a dog, if his master dies, might not eat food, which is a suboptimal decision.
This leads me into believing that one attribute of consciousness could be to make sub optimal decisions. But as argued by turing in his 1950 paper that "computers can very well make mistakes" leads me to conclusion that it cannot be a test for consciousness.
This leads me into asking how can we tell whether someone is consciousness or not. What can be a test or is it even possible to construct one? How can you know that I am even consciousness or not? The only reasonable answer(which actually can't be reasoned) is that soul implies consciousness.
So without this, how can I argue whether something is AI or not, which is not sentimentally driven but driven by arguments which can convince anyone in the world?
I am under the impression that the question is not well defined. I admire turing when he changed the question "do machines think', to something more concrete. I think a similar treatment is needed in this question too.