Turing test

A proposed test of machine intelligence in which a machine tries to imitate a human well enough to fool a human judge in conversation. Its outcome depends on conversational performance, not an inspection of the machine’s internal processes.

Imitation game: A conversational game in which an interrogator distinguishes participants by asking questions through written communication. Alan Turing framed his proposed test as a machine’s participation in this question-and-answer game.
Alan Turing: A British mathematician and computer scientist who made foundational contributions to computing and wartime codebreaking. He proposed the test in his 1950 essay on machine intelligence.
Winograd Schema Challenge: A proposed language-understanding test using sentences whose interpretation depends on commonsense knowledge. It targets commonsense reasoning that a conversational imitation may not reveal.
ELIZA: An early natural-language program created by Joseph Weizenbaum that simulated a psychotherapist through scripted text responses. Its conversational illusion showed how readily people could perceive understanding in a simple program.
Turing test protocols: Experimental procedures specifying participants, conversation limits, and how judgments of human identity are scored. Different protocols can produce different verdicts, so the test has no single universal procedure.
Computing Machinery and Intelligence: Alan Turing’s 1950 paper asking whether machines can think and proposing the imitation game as a substitute question. This essay introduced the test and explained why Turing reframed the original question.
Total Turing Test: A proposed extension requiring humanlike perception and physical interaction as well as conversation. It expands the original test beyond text-based conversation to perception and action.
Chatbot: A software system that conducts text or spoken conversations with users. Chatbots make the test’s central medium a routine form of human-computer interaction.
ELIZA effect: The tendency to attribute understanding or emotion to computer programs that produce convincing language. Judges may mistake fluent replies for human understanding when evaluating a machine.
Mind–body problem: The philosophical problem of how mental states relate to physical bodies and processes. The test offers a behavioral approach to questions about minds and machines.

Linked from 9 pages