Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Response by Ray Kurzweil to chatbot Eugene Goostman “passing the Turing test”

kurzweilai.net

Post reply on HN
Print view View original

101–110 of 164 posts

Re: Response by Ray Kurzweil to chatbot Eugene Goostman “passing the Turing test”

#101

Post by Paradigma11 »

macspoofing wrote: post #48
ambler0 wrote: post #31

I think most people misunderstand the intent of Turing's paper in which he describes his eponymous test. I agree with Chomsky's reading, which he laid out in "Language & Thought": "There is a great deal of often heated debate about these matters in the literature of the cognitive sciences, artificial intelligence, and philosophy of mind, but it is hard to see that any serious question has been posed. The question of…

>Fooling people into mistaking a submarine for a whale doesn't show that submarines really swim Why not? Presumably part of the act of fooling would involve 'swimming' submarines. I think passing a __strong__ Turing test actually does say something about our brains, cognition, and consciousness. And by a 'strong Turing test', I don't mean dinky 5 minute tests with a "Ukrainian boy". Imagine you carry on a 20 year rel…

I would guess that most contemporary philosophers do not fall into this traps but naturally they get no attention.

Why?

Journalists/ Interested parties ask philosopher about the mind/body problem, the matter of truth, god.... and if 9 out 10 answer that there is no real problem just misconceptions then the journalist will go to the 10th one who will get air time.

Re: Response by Ray Kurzweil to chatbot Eugene Goostman “passing the Turing test”

#102

Post by nardi »

thesteamboat wrote: post #80
computator wrote: post #70

Earlier quoted context omitted.

Can someone please help with the probability calculation? 3 judges and 4 participants What's the probability that any 2 or all 3 judges will pick a particular participant ("the computer") out of 4 at random?

Suppose that the participants are all indistinguishable. Each judge has a 3/4 chance of "being fooled". Let p be the event that the computer is correctly determined, and let q be the event that the computer is not selected. We calculate (1/4p+3/4q)^3 = 1/64(p^3 +3p^2(3q) + 3p(3q)^2+(3q)^3) = 1/64 (p^3 + 9p^2q + 27pq^2 + 27q^3). The probability, then, that the computer is chosen by random chance is 10/64 or approximat…

Which means in only 14 trials, there's a greater than 90% chance of passing the test. (Again, assuming the contestant is indistinguishable from the judges.)

Re: Response by Ray Kurzweil to chatbot Eugene Goostman “passing the Turing test”

#103

Post by fixermark »

drzaiusapelord wrote: post #28

The answers Ray got back are borderline embarrassing. They're designed to deflect the question by ignoring the request and asking another unrelated question or deflecting by humor. These are common ploys in faking AI chat. Hell, what 13 yo calls himself a "little boy?" I was in 8th grade at 13 and looking forward to high school. I can't imagine calling myself a "little boy" at an age when some of my peers were sexual…

Different tests are a good idea, but the devil is in designing them.

They'd have to both tell us something useful about the possibly-intelligent agent in question and not disallow anyone who we would consider an intelligent agent (more specifically, there's a whole suite of tests we could hypothetically use that would need to be discarded because they'd rule out entire sets of humanity as non-intelligence. Oops :-p)

Re: Response by Ray Kurzweil to chatbot Eugene Goostman “passing the Turing test”

#104

Post by coldtea »

balls187 wrote: post #55
robotresearcher wrote: post #47

Earlier quoted context omitted.

There's not much wrong with the University of Reading. Kevin Warwick, on the other hand is a complex character who errs on the side of hype and unjustified self-publicity much too often. He is not taken very seriously in academic AI and robotics, except when he causes a collective face-palm with claims like this. The press release people at Reading, or pretty much anywhere else, do not have the skills to assess wheth…

You might say he is the Ken Rockwell of Computer Science.

Yeah, if only Ken Rockwell was not entirely sane, and said things that make sense for the market segment he refers too.

If anything, it's the pixel-peepers and the "I need a bigger lens" that are non-scientific nuts (essentially gear fashion victims, who would buy a $10,000 Leica when a $1000 Nikon can give the same results).

Re: Response by Ray Kurzweil to chatbot Eugene Goostman “passing the Turing test”

#105

Post by DonHopkins »

Ben Shneiderman wrote an interesting essay entitled "Beyond Intelligent Machines: Designing Predictable and Controllable User Interfaces", that explained why he is strongly opposed to suggesting that computers are 'intelligent' or 'smart', and suggested better approaches to human computer interaction.

https://code.google.com/p/micropolis/source/browse/trunk/mic...

He wrote that back in 1992, and I think it's still very relevant now.

Limits to Imagination

I think we should have much greater ambition than to make a computer behave like an intelligent butler or other human agent. Computer supported cooperative work (CSCW), hypertext/hypermedia, multi-media, information visualization, and virtual realities are powerful technologies that enable human users to accomplish tasks that no human has ever done. If we describe computers in human terms then we run the risk of limiting our ambition and creativity in the design of future computer capabilities.

Re: Response by Ray Kurzweil to chatbot Eugene Goostman “passing the Turing test”

#108

Post by andrewla »

In thinking of this as a binary question (pass/fail) it feels like we're missing the point. As Kurzweil points out when quoting himself, "By the time there is a broad consensus that the Turing test has been passed, the actual threshold will have long since been achieved."

What we need is more of a Turing "score". The design would be a website (say) where each human participant will be presented with both "defend your humanity" challenges (as a subject) or "judge other's humanity". For the former, you'll try to convince an interviewer that you are human; for the latter, you'll be presented with two subjects, and asked to identify which is human (possibly with a "both human" or "both robots" option).

Based on this, individuals will get an ELO score (like chess) on how often they "win" the contest as a subject, that is, are identified as human (or as "more likely to be human than their opponent").

Computer programs will participate as subjects; the requirement will be that the behavior is deterministic given a random seed presented (to them only) at the beginning of the conversation,is to prevent cheating and allow reproducibility.

On an orthogonal level, participants acting as judges could be scored on how often they are correct in making the identification; and there's no reason that computer programs could not compete on this side as well. And this could even feed back into the score for humanity; you get more credit for fooling a good judge than fooling a bad judge, etc.

Re: Response by Ray Kurzweil to chatbot Eugene Goostman “passing the Turing test”

#109

Post by wlievens »

pizza234 wrote: post #84
macspoofing wrote: post #48

Earlier quoted context omitted.

>Fooling people into mistaking a submarine for a whale doesn't show that submarines really swim Why not? Presumably part of the act of fooling would involve 'swimming' submarines. I think passing a __strong__ Turing test actually does say something about our brains, cognition, and consciousness. And by a 'strong Turing test', I don't mean dinky 5 minute tests with a "Ukrainian boy". Imagine you carry on a 20 year rel…

There are a few problems with this line of reasoning. First, that there is an implicit assumption about how cognition and consciousness are defined, but there is no real definition. Following the example, it's not defined what a duck is, and "philosophers" actually don't try to elevate the question - they try to find answers to it. While a "strong" test for example would look appealing, the problem is that there is n…

The very hard problems you are referring to are called Strong AI. If your argument is that passing a proper Turing Test requires Strong AI, then, well, duh?

Re: Response by Ray Kurzweil to chatbot Eugene Goostman “passing the Turing test”

#110

Post by wlievens »

fooqux wrote: post #90
ambler0 wrote: post #66

Earlier quoted context omitted.

I think we agree. And I think Turing's point in choosing a conversation as the test was to pick something that would indicate the vast range of human experience. If a computer could pass a strong version of the test as you describe, then I would agree that we would have say, as you do, "if it looks like a duck, .... then it is a duck." But I don't agree that it renders the concept of conciousness meaningless (or at l…

I'm not so sure that just because you can fool me that an object is a duck, that it is a duck. At least when applied to general AI. Nor do I necessarily agree that just because I can have a conversation with an algorithm, that we should give that algorithm all the rights given to humans. For example, if I say mean things to that algorithm, even one that can fool me into thinking it's another human for decades, is tha…

> it would require surprising its creators to the point that they cannot fathom how it behaved the way it did

That requirement by itself is probably the easiest of th requirements.

Post reply on HN