Computing Machinery and Intelligence
In the essay Computing Machinery and Intelligence, Turing introduces a test, named imitation game, which can decide whether a machine is intelligent or not. In the test, there is a man who pretends to be a woman, a woman who tries to clarify that, and a machine that tries to be a man. If the interrogator, which object is to decide who the man is and who the woman is, fails to make the right decision, the machine should be considered intelligent. If a machine can talk to people for a long time and the people does not find out that this conversation is happening with a machine, it means that the machine can thinking. But Turing raises the concern that if a machine does “thinking”, but the “thinking” is different as what a man will do, the machine will still be considered not intelligent. The test may be too hard for machine. Turing thinks we should consider digital computers for two reasons. One is that the digital computer has already existed. Further more, the digital computers are basic machine, that is to say any digital machines can do the things that digital computers can do. Turing is sure that human will invent the thinking machine in the future. As for the thinking machine, it can talk to the human normally for most of the questions. In my opinion, in order to talk to human, the machine has been set up millions of tones and pieces of knowledge. It has the certain response to different questions. The difference between normal machines and thinking machines is that the normal machines only have certain response to certain questions. For example, if you ask a normal machine the same questions for many times, it would give you the same answers. You will find out it is the machine immediately since the boring response. However, the thinking machines are supposed to talking with the mood. It should talk by the tone similar to human talk. The response should be accurate as well as reasonable. Considering last example, a thinking machine should not