I’ve been curious about this question for a number of years. What is it that we need Artificial Intelligence for? The makers of Terminator 2 did a great disservice in cutting out the scene in which Dr. Miles Dyson explains to his wife why he is working on the neural network chip that is the heart of Skynet. He imagines pilots who never fly drunk and drivers who never fall asleep at the wheel. His dreams were more noble than the finished cut of the movie portrays.

There is a religious reason for pursuing artificial intelligence. You see, if you can show that intellect is just an algorithmic process performed on a vast quantity of data, then the creation of the human mind can be credited to the god of this world. There are people who believe that you can form artificial intelligence by feeding enough data into a lisp program. However, if you are feeding patterned data into a machine programmed to find patterns in data, even a stand up philosopher is not going to be impressed that you get some semblance of order out of it. Can intellect really be something that simple and yet be so rare in practice?

The chip in the terminator movie that I mentioned was a neural network. I wrote a paper on Neural networks once. They, like the organic brain, are very good at finding patterns. A network is trained by feeding in a pattern of inputs that connect to a pattern of outputs and the network alters its connections to match the pattern. This system is good at finding patterns that the programmer does not know exist. But, is that intelligence?

I am asking these questions because I do not believe that artificial intelligence is necessary. It may not be possible, yet that is not an answer to the question. As I see it, these words on your screen appear intelligent. These words came out of a word processing program that I have used since I really wrote something memorable for the first time. This word processor is not intelligent in itself.

This is what I call virtual intelligence. If we can maintain true intelligence in the world, then we can create our machines as an amplification of our intellect as our other tools amplify our physical abilities. Virtual intelligence appears intelligent because it is reacting to true intelligence. In this respect, virtual intelligence is a series of reflexes. Can we not achieve every promise of artificial intelligence with a proper application of virtual intelligence?

This entry was posted on Monday, July 9th, 2007 at 12:00 am and is filed under Mulligan Adventures. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.