Turing test (UNCLASSIFIED)

[Mullins, Bill AMRDEC]

Classification: UNCLASSIFIED
Caveats: NONE

The Turing test is a situation where a person communicates back a forth
with what may be a person, or it may be a computer.  If the person can't
reliably distinguish the computer from the person, based on the content
of the messages sent back from the computer/person, then the computer
passes the test -- it is "artificially intelligent."

In a response to a comment on a Language Log post
http://languagelog.ldc.upenn.edu/nll/?p=1701

Geoff Numberg said
"Which only goes to show that the Turing test can work both ways: do
something dumb enough, and it's hard to tell you from a machine."

This strikes me as similar in form to Niven's law ("Any sufficiently
rigorously defined magic is indistinguishable from technology."), an
inversion of Clarke's Third Law ("Any sufficiently advanced technology
is indistinguishable from magic."

Did Nunberg come up with a new law?  Has anyone else ever stated the
idea that way?



Classification: UNCLASSIFIED
Caveats: NONE

------------------------------------------------------------
The American Dialect Society - http://www.americandialect.org