The Turing test is named after World War II mathematician and codebreaker Alan Turing – you might have seen the 2014 film, ‘The Imitation Game’, starring Benedict Cumberbatch (BBC’s Sherlock, and Marvel movies’ Dr. Strange), playing Alan Turing. The science of artificial intelligence emerged from Turing’s ideas.
Turing’s test is a way to explore if a machine can answer any question put to it. The machine has to be hidden from view or hearing, in a different room is better. Then a human interrogator can ask any questions by typing text, and based on the text-based answers, decide whether they are talking with a human or a machine.
When you are in the ‘I’m a Scientist, Get me Out of Here!’ chatroom talking with Scientists you could say you are an Interrogator and you are judging whether you are talking to a real scientist or a pretend one!
The Turing test is about indistinguishability – can you distinguish human from machine? Although Turing’s idea is based on a human language test, the indistinguishability measure can be used anywhere where a machine is designed to do a job once done by a human. For example, driverless cars, pilot-less planes. If the driverless car, or pilot-less plane is as good as a car driven by a human, or a plane flown by a pilot then you can say the machine has achieved human-level performance, its capability is indistinguishable from a human doing the same task.
You can see pictures of human language Turing test experiments I’ve conducted on these blogs:
If you have any other questions about the Turing test, or you would like to read more, you can read my paper, ‘Turing’s Misunderstood Imitation Game and IBM’s Watson Success’ at the link below:
Comments