The Turing test was developed in 1950 by Alan Turing as a way of determining the ability of machines to exhibit behaviours that are indistinguishable from or equivalent to the intelligence of humans.
2. The Turing test was developed in 1950 by
Alan Turing as a way of determining the ability
of machines to exhibit behaviours that are
indistinguishable from or equivalent to the
intelligence of humans.
3. Matthew Ledvina 3
HUMAN EVALUATION
The basis of the Turing Test involves
a human evaluator listening to con-
versations between another human
and a machine, designed to respond
as a human would. The evaluator is
aware that one party is a machine
but not which one. The machine is
deemed to have passed the test if
the evaluator cannot reliably deter-
mine which party is which.
TEXT-ONLY COMMUNICATION
To ensure fairness in testing, the
conversation between the two par-
ties takes place in a text-only for-
THE TURING TEST DOES NOT REQUIRE THE MACHINE TO ANSWER
QUESTIONS POSED CORRECTLY. THE RESULTS ARE BASED ON HOW
CLOSELY THE ANSWERS ARE TO THOSE A HUMAN WOULD NATURALLY
RESPOND WITH, NOT THEIR ACCURACY.
mat. This means the machine’s
intelligence of responses can be
evaluated without depending on its
ability to render those words as hu-
man-like speech.
NO CORRECT ANSWERS
The Turing Test does not require the
machine to answer questions posed
correctly. The results are based
on how closely the answers are to
those a human would naturally re-
spond with, not their accuracy.
Artificial intelligence has progressed
hugely since the Turing test was de-
vised in 1950.
4. Visit the blog of
Matthew Ledvina
to read about Yuval
Noah Harari’s
book Homo Deus,
which explores the
implications of the
digital age.
Matthew Ledvina