Turing Test
Turing Test
A test method used to determine whether an AI is as intelligent as a human.
In Simple Terms
The Turing Test is an experiment that checks whether a machine has intelligence similar to a human's. In the experiment, a human judge chats with someone using only text messages and tries to guess whether they're talking to a human or a machine. If the judge mistakes the machine for a human, the machine is considered to have intelligence. It's often cited in discussions and research as an important benchmark for thinking about AI intelligence.
Behind the Name
The name comes from Alan Turing, the brilliant British mathematician who proposed this method. It's one of the widely known tests in the history of AI — a long-standing method for exploring whether a computer has intelligence.
Take a Closer Look!
The Turing Test is an experimental method for determining whether a computer or other machine can think like a human.
It's one of the representative ways of testing this idea, discussed since the early days of AI's history.
In the actual experiment, there's one human judge, while a human and the machine being tested are set up together in a separate room.
The judge can't see either of them and instead has free-form conversations with both using only text messages.
There are no restrictions on what can be discussed — the judge is free to ask anything.
After the conversations end, the judge guesses which one was the human and which was the machine.
If the judge mistakes the machine for a real human, the machine is said to have passed the Turing Test.
The idea is that if something can talk naturally enough to be indistinguishable from a human, it can be considered intelligent.
Broadly speaking, this is a way of judging intelligence based purely on visible behavior, not on what's happening inside the other party's mind.
This idea — judging intelligence by behavior alone — has become one of the foundational ways of thinking about AI intelligence.