Since the advent of computers, science fiction writers have come up with plots with intelligent machines that capture the world and make people of slaves. Scientists at first laughed at this, but as information technology developed, the idea of a reasonable machine ceased to seem so incredible. To test whether a computer can have intelligence, a Turing test was created, and it was invented by none other than Alan Turing, whose name this technique was named. Let's talk in more detail about what kind of test this is and what it actually can.
How to pass the Turing test?
Who invented the Turing test, we know, but why did he do it to prove that no machine is like a man? In fact, Alan Turing was engaged in serious studies of "machine intelligence" and suggested that it is possible to create such a machine that can carry out mental activity like a human being. In any case, back in the year 47 of the last century, he stated that it is not difficult to make a machine that could play chess well, and if it is possible, then it is possible to create a "thinking" computer. But how to determine whether the engineers have achieved their goal or not, does their child have intelligence or is it another advanced calculator? For this purpose, Alan Turing created his own test, which allows us to understand how much the computer intelligence can compete with the human.
The essence of the Turing test is the following: if the computer can think, then when talking, a person can not distinguish the machine from another person. The test involves 2 people and one computer, all participants do not see each other, and communication takes place in writing. Correspondence is conducted at controlled intervals so that the judge can not determine the computer, being guided by the speed of replies. The test is considered passed, if the judge can not say with whom he is in correspondence - with a person or a computer. To complete the Turing test has not yet been possible for any program. In 1966, Eliza's program managed to deceive the judges, but only because she imitated the techniques of a psychotherapist using a client-centered technique, and people were not told that they could talk to the computer. In 1972, the program PARRY, imitating a paranoid schizophrenic, was also able to deceive 52% of psychiatrists. The test was conducted by one team of psychiatrists, and the second read the transcript of the recording. Before both teams was the task to find out where the words of real people, and where the speech program. It was possible to do this only in 48% of cases, but the Turing test involves communication in on-line mode, rather than reading the records.
Today there is a Löbner Prize, which is awarded according to the results of the annual contest to programs that were able to pass the Turing test. There are gold (visual and audio), silver (audio) and bronze (text) awards. The first two were not awarded yet, bronze medals were given to programs that could best simulate a person during their correspondence. But this kind of communication can not be called full-fledged, since it more closely resembles a friendly correspondence in a chat, consisting of fragmentary phrases. That's why
Inverse Turing test
One of the interpretations of the inverse Turing test was faced by everyone - it's annoying requests of sites to introduce captcha (CAPTHA), which are used to protect against spam bots. It is believed that there are not enough powerful programs yet (or they are not available to the average user) that can recognize the distorted text and reproduce it. Here is such a funny paradox - now we have to prove to computers our ability to think.