Failing the Turing Test

Friday, June 13, 2014

Transcript

It was widely reported this week that for the first time ever, a computer program had passed the "Turing Test." The trouble is, the story was a sham. Brooke talks with Tech Dirt's Mike Masnick about how the media should have known better. 

Guests:

Mike Masnick

Hosted by:

Brooke Gladstone

Comments [4]

Ryan from MA

Thank you, On The Media, for covering this. It was really bugging me.
One additional thing to note, though, is that this chatbot did *not* pass the Turing Test as Alan Turing defined it. Fooling 30% of the judges, or whatever, was what Turing predicted a computer would be able to do in 50yrs (from the time he wrote the paper).

Here's a link to a discussion of the story on Leiter Reports (a Philosophy blog): http://leiterreports.typepad.com/blog/2014/06/a-computer-that-passed-the-turing-test.html#comments

Here's David Chalmers on it:

>>"The chatbot didn't pass the Turing test. In Turing's original article, he predicted that in fifty years, machines would be able to fool 30% of judges into classifying them as human after 5-minute conversations. The organizers have somehow bamboozled the media into taking this prediction as the criterion for passing the test. In Turing's original article it's quite clear that it's nothing of the kind. In fact, in a follow-up discussion he says that he doesn't think the full test will be passed for at least a century. It's also worth noting that the bar has been lowered considerably by having the chatbot pretend to be a 13-year-old with English as a second language. If we're allowed to lower the bar like this, one can trivially write a Turing-test-passing program whose responses are indistinguishable from a human who is asleep!"

Jun. 17 2014 05:03 PM
Raymond Parsick from New York

Is it true that Turning mentioned as the ultimate test of distinguishing a computer from a person would be that the person would show some extra-sensory perception in guessing random cards, getting maybe 40% correct while the computer would get nearer to 33% by chance?

Jun. 16 2014 10:34 AM
Robert Thomas from Santa Clara

The Turing Test is indeed a real thing, discussed among those interested in human cognition and in the modeling of cognition. As Ventifact notes, Turing proposed it (among other tools) as a sort of thought experiment, as a way to steer discussion.

Foolish persons, perhaps wishing to play upon Turing's reputation and on those who latterly recognize his name, the tragedy of his personal life and injustice visited on him, have promoted the TT as some sort of ultimate challenge for machine modeling. It's not and has never been considered so.

But why should "the media have known better"? The popular media is generally UTTERLY INCAPABLE of usefully or sensibly interpreting the advent of ANY TECHNICAL TOPIC, of any kind.

Jun. 15 2014 05:44 PM
Ventifact from Raleigh, NC, USA

Those scare quotes around "Turing Test" don't belong in the story summary. The Turing Test itself is a real thing, although perhaps most real in the form of a thought experiment. Putting quotation marks around the phrase gives the impression that the Turing Test itself was made up as part of this sham story.

Jun. 15 2014 05:09 PM

Leave a Comment

Email addresses are required but never displayed.