Artificial Intelligence - Essay Example

Artificial Intelligence In 1951, Alan Turing proposed the test known as ‘The Imitation Game’ in a paper trying to settle the issue of artificial intelligence. In the first test, he explained his arguments with no application of computer intelligence. Imagine you have three separate rooms, which are connected via computer screen and keyboards to each other. A man sits in one room, a woman in the second room and a judge in the third one. The work of the judge is to identify the man between the two people talking to him through the computer. The man is trying to fool the ‘judge’ that he is not a male. The woman, on the other hand, attempts to deceive the ‘judge’ so that he can identify her as a male (Moor 142). Alan Turing, then, modified the ‘imitation game’ and put a man or woman, and a computer in another separate room. The judge’s job was to decide who is a man or machine among the two contestants. Based on the second version of the experiment, Turing argued that if the ‘judge’ can pick a computer or a human using his mind, then, the computer is a simulation of a human and hence it is intelligent. The Turing has been modified further in that the imitation game has changed where there is a single contestant, a human or a computer. The judge’s job in this test is to decide whether the single contestant is a human or a machine (Moor 143). Alan Turing proposes that the definition of intelligence can be passed to machines (computers) in that he believes that computers can be programmed to act like humans. He advocates for the question and answer method because it includes the human behaviour of talking by responding to queries. In his 1951 paper, Turing writes that “in fifty years’ time, computers can be programmed to play the imitation game such that any average interrogator (judge) will have less than 70% chance of identifying the correct contestant after asking questions for five minutes.” The Turing test supports the fact that if a human being is replaced by a computer and the game’s results remain unaffected, then, the machine would be equally intelligent as a human and capable of thought (Moor 144). Turing supports his arguments by stating that machines are capable of thought if only they cannot be distinguished from humans solely on the basis of interaction via “written” messages, such as typewritten questions, answers and teletypes. Turing supports this definition by arguing that written interaction are considered as adequate means of evaluating intelligence of a person (Moor 145). He addresses the possible objections to his view on intelligence by simply appealing to the critics to accept that a formal system is the most effective way of capturing intelligence. In other words, human intelligence is obviously measured using the formal interactions, for example, the question and answer method used in the Turing tests. Although there are topics of extrasensory perception and romantic love that define intelligence differently, Turing believes that his tests are adequate for assessing intelligence. Critics of the Turing Tests The Chinese Room experiment was developed by John Searle in 1980 to counteract with Turing’s arguments. Searle proposes that computer have no capability of thought as Turing argues in his tests. In his experiment, Searle uses the question and answer method to demonstrate how Turing’s conclusions are invalid. On one hand, we have native Chinese speaker and on the other hand, a monolingual English speaker. The Chinese speaker’s job is to ask the English speaker in Chinese language. The English gets the answers by hand tracing the program of Chinese by following instructions given in English. Searle concludes that he has made the interaction formal and created sufficient chances of depicting the human behaviours, but, the monolingual English speaker has not understood anything (Yang 29). Searle rejects the computational functionalism theory that is used to apply the operational definition of intelligence in human beings to machines (computers). The computational theory of mind suggests that mental states are similar to the software of a computer. In other words, your mind is the software and your brain is the hardware. Searle also disregards Turing’s claims that computers can one day think like human beings (Yang 30). He argues that using the digital computers you cannot run classical artificial intelligence program. He believes that machines can understand the meanings of some words and sentences. Turing tests have also received rejections and criticisms from philosophers who claim that the tests are not adequate to assess the intelligence of human beings. Thus the nature of intelligence in humans cannot be applied to machines. The questioner in the Turing experiments interacts with the contestants through language, which is inadequate in capturing all types of human intelligence. In response to these oppositions, the advocates for Turing tests claim that the tests for general intelligence and not all sorts of intelligence. The Chinese Room experiment has also been rendered invalid in support for the Turing tests. Philosophers argue that Searle’s experiments cannot be performed in this world (Yang 34). Therefore, despite John Searle disapprovals of the Turing tests, he also receives criticisms that nullify the validity of his experiment and arguments. It could be true that Turing tests are not adequate in measuring intelligence since many theories have disapproved Turing tests significantly since 1950. For example, the toe-stepping game, which was similar to the Turing tests, was rejected, as well. It was concluded that the toe-stepping game only focused on the ability to step on toes and assumed other human abilities. Researchers in artificial intelligence have ascertained that the Turing Test fails to capture most important indicators of intelligence and hence his arguments are invalid. Generally, Turing tests are based on one aspect of intelligence and hence we have many artificial intelligence indicators ranging from motor skills, communication skills and sensory thoughts. In summary, Turing tests are controversial among many philosophers of artificial intelligence. Then, the other AI tests put across to counter claims that computer can think have also been criticised significantly. It, therefore, means that it has not been discovered how computers can behave like human beings while capturing all aspects of human intelligence. Works Cited Moor, James H. The Turing Test: The Elusive Standard of Artificial Intelligence. Dordrecht [u.a.: Kluwer, 2003. Print. Yang, Xin-She. Artificial Intelligence, Evolutionary Computing and Metaheuristics: In the Footsteps of Alan Turing. Berlin: Springer, 2013. Print. Read More