Can Computers be "intelliegnt?" Can a computer "think?"

"She is intelligent." You can interpret that sentence in many different ways. Some people would interpret this as "she thinks fast," while others will take this is "she knows a lot." Personally, I find both of these definitions somewhat lacking. Someone can know a whole lot but be incapable of putting things together in an innovative fashion; others could think fast, yet be extremely stupid. A more satisfactory way to define it is "someone who can adapt their actions to a particular situation." Of course this is overly generalized, and therefore might not prove satisfactory to some people.

There are many important aspects when it comes to human intelligence, which I believe most people could agree on, such as: the ability to use intuition, common sense, creativity, judgement, logical reasoning, and a strong base of knowledge. While human intelligence is powerful, it has it's limits. For instance, humans have a limited knowledge base. Another example is that humans process information of a serial nature much slower than the computers we have today. Human intelligence is not infallible.

Two ways that people demonstrate their intelligence is by communication, and by demonstration of previous knowledge.

In order to communicate effectively, one must be able to synthesize a message in such a way as to express one's meaning to his intended audience. Doing this usually require some judgement calls by the person sending the message, such as predicting the level of sophistication of a recepient. Likewise, understanding a message also requires intelligence. A listener must know most of the words a speaker is using, and he must have some basis of knowledge about the content that a speaker is talking about. Additionally, the listner must then pull everything that the speaker has been talking about together, and put it in context.

Next we must define learning. For the purpose of what we are talking about, we will define it as the ability to adapt one's behavior to new situations. The ability to learn is generally considered a vital component of intelligence.

Now that we have broadly defined intelligence, we can go on to our next question: Can a computer be intelligent?

Philosiphical and ethical questions also arise. People often feel threatend by the possibility that a machine can think. This has been heavily debated for a long time.

Obviously, a computer on it's own would be incapable of thinking. A program must run to tell the computer what to do. So how do we diffrentiate between an "intelligent " program from a "non-intelligent" program? If you were to stay with a strict, dictionary definition of intelligence, then you could conceivably argue that all programs are intelligent, since computers are exceptionally well suited for apprehending facts, propositions, and relations. However, is it possible for a program to reason with these facts? If you define manipulating data as reasoning, then yes, most all programs are intelligent. But obviously not all programs can be intelligent. Here's an example: A database program does not think the same way a person thinks. However, when a file clerk does the same task it does require intelligence. Hence, a paradox arises.

Here's another example. If you could get your dog to fetch your newspaper, you might call it intelligent. But I doubt it would be too difficult to construct a robot programmed to handle the same task. However, most people will say the robot isn't smart because it is simply doing a task, not thinking about it. Also, many people will say that this isn't an intelligent program because they can understand what the robot is doing.

There is also the question of free will. Thought has always been connected with the concept of free will. Only an entity with the will to think can think. The question "Can a computer choose to do something" is hotly debated. For example, in a game of chess, a computer evaluates many possible combinations of moves, and picks the best one. Did it not just choose the move? But it was programmed to do so, so did it really have a choice?

There are compelling arguments on both sides. What do you think about it? Sorry if I have presented this information in a lacklustered manner. It is late and I am sleepy. I will edit this thread tomorrow.