The Turing Test

Is the Turing Test a sufficient test? That is, if a machine passes the test, would you agree it is intelligent?

If intelligence could be defined only as the ability to solve problems, then I believe a machine that passes the Turing Test could be considered intelligent. But I do not believe this definition is sufficient in describing intelligence. Just because a machine passes a test, merely by answering questions via a mechanical link, in such a manner that a human cannot distinguish the machine from another human, does not mean it is intelligent. It seems to me that passing the test would be more indicative of the machine’s ability to mask the fact that it is a machine.

Perhaps this could be compared to a female who has had surgery to remove her breasts, to implant an artificial penis, and has taken hormones to deepen her voice and to grow more facial hair. This female might give the impression of being a male. In spite of appearing as a male, she is a female, at least according to the biological definition of her gender at birth. Even if she went through the motions of acting like a male, would that mean she is male?

In my opinion, this is a matter of semantics. Perhaps new vocabulary words need to be invented that could more specifically address the nebulous boundaries of a word such as “intelligence”, when attributed to inanimate objects such as computers. And for that matter, such words as “male” and “female”, when attributed to individuals who experience gender identity issues.

Is the Turing Test a necessary test? That is, does a machine have to pass this test in order to be intelligent?

I do not believe it is necessary that a machine pass the Turing Test in order to be considered intelligent. For example, a computer is able to solve problems and one might consider this to be intelligent, based on the definition of being able to “solve problems”. A machine that can answer questions in such a way as to deceive someone into believing the machine is human is, in my opinion, not necessary for determining whether or not the computer is intelligent.

Will a machine ever pass the Turing Test? Why or why not?

I do not believe a machine will ever pass the Turing Test. This test requires that an interrogator question a hidden entity, which is either a computer or another human being. The questioner must then decide, based solely on the hidden entity’s answers, whether he had been interrogating a man or a machine. Although a computer could respond to questions, I do not believe the computer could think like a human being. I think we saw a little bit of what this would be like when we explored the A.L.I.C.E. website. Although the computer could appropriately answer many of the questions, there were times when the syntax of the computer’s responses were not appropriate for the questions. Or there were times when the computer was not programmed to answer the question and tried to hide its “ignorance” by changing the subject, not answering the question, or by asking its own question. I am not saying a human being wouldn’t do the same thing. But given enough opportunities to receive an answer from a computer, I feel that, eventually, the interrogator would be able to figure out that the computer was not a human. In other words, at some point in time, the computer would just come across as fake or unreal.

Will a machine ever be intelligent? Why or why not?

Again, this depends on the definition of “intelligent”. If intelligent means “thinking” as a human, then, no, I do not think a machine will ever be intelligent. I believe a machine’s intelligence is simulated, not real. Simulated actions which might be interpreted as “thinking” or “intelligent” are a result of the human minds that programmed the computer. If one defines intelligence as including information processing, memory, storage, retrieval, assimilation, and synthesis, then a machine might be described as being intelligent. However, if consciousness is a part of intelligence, then machines, being man-made, could not be described as intelligent.