Conceptica

Artificial Intelligence - A philosophical view

Now a days we see a great interest in the field of Artificial Intelligence(AI). AI together with Nano-technology is predicted to be the most transforming technologies of the 21st century. But the truly interesting problem is “do we really understand what AI is?” There’s so many views on this. Most noticeable view point is that AI is, improving computer algorithms to a point where these can solve whatever the problems that humans can solve and more. So far what we see is that this view can lead to a lot of domain specific algorithms and programs which can be loaded on to a powerful computer and once you have enough programs in it, it should be more intelligent than a human being. There’s many algorithms developed already like neural networking, genetic algorithms etc. These algorithms have achieved many amazing feats in natural language processing, image recognition and so forth. But we have to really take a step back and think, will this algorithmic evolution truly bring about actual intelligence in an artificial form? I have a few other questions as well,

1. Will an algorithmically run machine ever be able to conceptualize and think out of the box about solving any serious problems that humans face today? The greatest power of human intelligence IS its ability to innovate over and invent concepts(even invent artificial intelligence).
2. Another huge question mark is general intelligence. Will the same algorithmically driven AI be able to dive in quantum mechanics in the day and take good care of a child at night? Some will say we don’t need this kind of intelligence. But I think it’s an integral part of being intelligent.
3. Normally we find a problem and we set goals for an AI to achieve. Can’t we have goal-less(not predetermined) AIs, Which will just do whatever it wants and achieve whatever it wants? For human intelligence this is the norm.

So how do we get over all these difficulties and achieve truly INTELLIGENT artificial intelligence?(assuming it’s actually possible) This what I’ll address in my next bog post where I’ll disclose my own concept of artificial intelligence.