What Does Turing Test Mean?
A Turing test is an evaluation introduced by Alan Turing in 1950 to determine whether or not someone can tell if they are communicating with a computer program after five minutes of conversation.
Turing tests are controversial because while the mathematician Alan Turing was interested in the idea of whether or not a machine could be sentient and think for itself, the test he proposed does not do that.
Instead, Turing's test, which is carried out through an activity called the Imitation Game, can only be used to evaluate whether a program's natural language generation (NLG) capabilities are strong enough for someone to think that computer-generated responses came from a human being.
Techopedia Explains Turing Test
In the 1950s, mathematician Alan Turing released a paper entitled "Computing Machinery and Intelligence." Turing was interested in the idea that under the right circumstances, a machine could think like a human being. He proposed modifying an activity called the Imitation Game to gather data and hopefully, answer the question.
How Turing's Imitation Game Works
The Imitation Game was a popular parlor game in the Victorian Era. It required one person (called the interrogator) to send the same questions to a man and a woman in another room, and then guess which set of responses came from the woman -- and which came from the man.
A more modern version of the game requires two human players and a computer. Just as in the original game, one human (the interrogator) is put in a separate room. The interrogator then has five minutes to text each of the entities in the other room questions and guess which set of answers was generated by a computer.
To ensure fairness, the test is repeated multiple times with different interrogators. At the end of the test, if the jury of interrogators believe they are communicating with another human being at least 70 % of the time (when they have actually been talking to a computer program) the software's creators can legitimately claim their AI programming has passed the Turing Test.
In 2022, two artificial intelligence chatbots -- Google’s LaMDA AI and OpenAI's ChatGPT -- successfully passed the Turing test.
One of the first computer programs that could simulate human conversation in a somewhat convincing way is called ELIZA. Although this chatterbot was not able to pass the Turing Test, it did inspire the phrase Eliza Effect, a term used to describe the willingness people have to attribute human qualities to a computer program. The phrase if often used as a synonym for personification in the context of information technology.