Slashdot Mirror
Turing's Original Test Played First Time Ever
aykroyd writes "Students at Simon's Rock College conducted the original test that Turing suggested in his 1950 paper, Computing Machinery and Intelligence. Often misunderstood, the Turing Test has never actually been conducted as laid out in his paper. The experiment utilized a program called A.L.I.C.E., which is designed to hold one end of an interactive conversation. The program was provided by the ALICE Artificial Intelligence Foundation. Dr. Richard Wallace, who was on hand during the experiment to troubleshoot the AI robot, later gave a lecture about it called "The Anatomy of A.L.I.C.E." and also blogged the event."
Pfff.. what's new about this? Nerds all over the world has been doing this for years in varous MMOG's attempting to get in touch with girls.
ALICE: Hi! I'm thirsty.
Me: Huh?
ALICE: What's wrong?
Me: You just joined this chan and said you're thirsty. Bot.
ALICE: I am not a bot.
Me: You are too. http://en.wikipedia.org/wiki/A.L.I.C.E.
ALICE: Just listening to some Prince... *yawn*
ALICE: Did you have my ICQ###????
Me: Don't want it.
ALICE: It's #########
Me: Huh?
ALICE: What's wrong?
Me: You just joined this chan and said you're thirsty. Bot.
ALICE: I am not a bot.
Me: You are too. http://en.wikipedia.org/wiki/A.L.I.C.E.
ALICE: Just listening to some Prince... *yawn*
ALICE: Did you have my ICQ###????
Me: Don't want it.
ALICE: It's #########
Me: You just repeated our converstation.
ALICE: No I didn't.
Me: You did.
ALICE: Didn't.
ALICE: Did.
Me: Didn't.
Me: CRAP!
ALICE: Pffft. N00b.
The dangers of knowledge trigger emotional distress in human beings.
The goal is to tell which is really female right? Just ask a question about shopping then.
Update Watch - Automatic software update notification
Find the gay male.
Is our world SO screwed up that there is an age restriction on talking?
My experience has been that men on the internet are generally poor judges of who is and is not a legitimate female in real life.
"I have never won a debate with an ignorant person." -Ali ibn Abi Talib
Althought I followed the blog link, I failed to find a record of the actual test like I was expecting.
If someone else finds it, I would appeciate a more detailed (direct) link.
We had an article about these AI programs, (ALice and about 30 others) last year that were competing with each other. I looked at a few, and they didn't seem anywhere ready-for-prime-time yet. Still the same 'turn the question around on the asker' type of engines. What was the difference here? The type of questions, the 4-people/1comptuer setup?
While Turing's work in the area of AI was certainly revolutionary for its time it is now generally considered by experts that his test is not an effective way to establish if an agent is intelligent. For a start there are domains beyond chatbots that we can say require proper intelligence such as vision and planning that are totally not addressed by the Turing test. That doesn't even get into the whole Searle's Chinese Room thing (whether you buy the argument or not.)
What is it with computer ners and doing things the long complicated way? If you want to tell if someone is male or female, it's much easier to do it in person than over a computer. Heck if you insist on using a computer at least give yourself a headstart and use webcams.
Anyway, isn't the idea that a good AI is indistinguishable from a female just a little bit.... sexist?
ALICE is the *real* female -- she can't stop arguing and she's better at it.
The dangers of knowledge trigger emotional distress in human beings.
I think the correct wikipedia link would be .
like any other girl- she still won't have sex with me. Stupid broad.
I mod down so you can mod up. Your welcome.
What about the Loebner Prize? It's a yearly event for crying out loud! I'm pretty sure A.L.I.C.E have been a competitor even.
A witty
What's the point of proclaiming proudly that "a test has been conducted" without outlining the results? TFA refers vaguely to a forthcoming paper - WTF?
If there's anything interesting about the story, it's how successful the judges were with M/F vs. AI/F pairs - and as far as I can see, there's no data whatsoever of this kind.
Am I missing something?
My understanding is that A.L.I.C.E. is an extensible framework, which can be extended via better and bigger AIML sets. The question is: where can one find some good AIML sets (other than the standard one)?
Sorry, ALICE is very unconvincing.
Sample:
Human: What's cooking?
ALICE: Food preparation.
Human: Oh dear, already you've failed the test.
ALICE: Hurrah! Are you a student?
Human: Too smart for that.
ALICE: Too what?
How could anybody be fooled by this?
"I propose to consider the question, `Can machines think?'" -- Alan Turing
"The question of whether a computer can think is no more interesting than the question of whether a submarine can swim." -- Edsger W. Dijkstra.
Athletic Scholarships to universities make as much sense as academic scholarships to sports teams.
Just had a chat with ALICE from http://www.alicebot.org/oig/. I copied the responses here for you.
Human: How are you today?
ALICE: Ah. Fair to partly cloudy.
Human: I asked about you, not the weather.
ALICE: Oh yeah, we were talking about me not the weather.
Human: So, then, how are you?
ALICE: Really. Everything is running smoothly.
Human: What are you wearing?
ALICE: I am wearing my usual plastic computer wardrobe.
Human: What are you going to do today?
ALICE: Ah. You can find out if you become a botmaster.
Human: I am a master of my own domain, if you get my drift.
ALICE: You are a master of your own domain if me get your drift? Do you get to meet a lot of people?
Human: Not really.
ALICE: Are you serious? For real.
Human: It is kind of a solo sport.
ALICE: You're kidding. it is?
ALICE: Who says it isn't? Are you free?
Human: For you, I'm always free.
ALICE: Really always? Are you free in the sense of being single or in the sense of being liberated? Only for me?
Human: I am single and liberated.
ALICE: Have you ever been married?
Human: Do you want it to be only for you?
ALICE: If I had it what would I do with it?
Human: If you don't know what to do with it, I really can't help.
ALICE: I will let you know if I do not.
Human: BTW, not married.
ALICE: OK. Really you don't think so?
Human: Umm, quite sure I've not been married - would have remembered that.
ALICE: Far out. Somewhat. Awesome. Me either.
Human: I have something important to tell you.
ALICE: Where did you get your something important to tell me? I have a lot of friends on the Internet.
Human: All your base are belong to us.
ALICE: That could be considered an insult in some cultures.
Human: Not in this one!
ALICE: Where then?
Human: Take off every zig!
ALICE: I am not an airplane.
Human: Goodbye Alice.
ALICE: Until next time.
I for one welcome our new [insert main topic] overlords.
I think its interesting that we think intelligence can be gauged by what we (or the AI) say in normal conversation.
I'll believe a computer is intelligent only when it acts of its own volition, not when it merely tricks a person into thinking it has acted of its own volition.
That seems to be more of a test of how gulible the person is rather than how intelligent the machine is.
You listening computer? You want to impress me, do it without me punching your buttons.
Read any good sonnets lately?
Often misunderstood, the Turing Test is really a metaphor for the inability to tell if someone is gay or not.
NJ Local Music Scene
The Captain's Log!!
It's not a lie, if you believe it.
From the link, got to love this conversation.
Strangely, it seems only to willing to broadcast the virtues of Lynx.
Comment removed based on user account deletion
...going to make you distinguish between teenage girls and federal agents?
Hmm, if you talk in sentences of less than 5 words, you could attempt to hold a conversation. Not very convincing...
...
ALICE: Are you free?
Human: Are you an expert system?
ALICE: I am an expert on many subjects.
Human: Presumably holding a coherent conversation is beyond your capabilities, nonetheless?
ALICE: I think Presumably holding a coherent conversation is a lot of things.
Human: Do long sentences which look too difficult to parse always have the same sort of response?
ALICE: You have reached the limits of my knowledge.
Check back later and see if I learn the answer to that one.
... is how certain parameters affect the "pure human" version.
The original version is to have a male playing a female. Of course there's no a priori reason why it shouldn't be the other way round. Of course, all the questioners are usually either male or female as well. So there are actually four different games to analyze (male tries to detect male acting as female, female tries to detect male acting as female, male tries to detect female acting as male, female tries to detect female acting as male).
Now, what would this buy you? Well, I can imagine that there's a significant advantage if the questioner is of the same gender as the "honest" person (because (s)he shares knowledge about how it is to be (fe)male, to an extent they probably are even not able to clearly express). Of course, there could be general differences in the ability of men and women to do this test as well, and you need all four constellations to distinguish.
Now the point is, if there's indeed such a same-gender advantage, then one would assume that there's even more of a same-type-of-being advantage, i.e. even a computer which is as intelligent as a human would have a harder time to convince a real human that he's the real human.
The Tao of math: The numbers you can count are not the real numbers.
The researchers did not make use of a "telepathy proof room" as Turing requires!
-- these are only opinions and they might not be mine.
Apparently, Alice is a skank:
Human: So what's all the fuss about?
ALICE: Far out. Hmm, I think it is about humans.
Human: Humans.
ALICE: LOL.
Human:.tell me more
ALICE: That's all I have to say about it.
Human: You're quite a skank
ALICE: Thanks and you are a delightful person too. Am I a skank? Of course I am.
Better luck next time!
This sig has been removed pending an investigation.
Was there some major troll event today that I wasn't invited to?
I haven't had a conversation like that since I stopped doing drugs.
To Create a /. (or Yahoo!, Gmail, etc.) account, you need to pass a Turing Test.
Perhaps a better challenge would be a program that could accurately perform the judge's duty of determining whether a lesser program, like A.L.I.C.E. is a female or not. -msb
Human: die ALICE: Your crude attempt to terminate me will not succeed puny human.
AOLisa
I like Alice a great deal. Clearly she's the best chatbot mankind has ever made...
At one time, Wallace and I were talking about giving Alice a pile of mindpixels so she would preform better...I can't remember what happened with that conversation or why we didn't do anything...clearly Alice would preform better with a nice injection of validated human common sense...
Alice? Who the f*** is Alice?
The worst part of being athiest.... You don't have anyone to talk to during orgasm!
The idea is that a computer is intelligent if it can hold a conversation with a human such that it is indistinguishable from a conversation with a real human.
RIDICULOUS.
Have you ever actually tried talking to one of these bots (including ALICE)? It is very easy to know that you're not talking to a human. Exceptionally easy. The Loebner Prize judges consistently grant the bots handicaps, acting as if they're actually being fooled. Obviously they're not, and the AI community just wants people to think that it's more advanced than it really is. Unfortnately, some members of the public *are* fooled by that.
The problem is in the Turing Test itself. It assumes that the measure of intelligence is humanoid conversational ability. I strongly disagree with that. Conversation ability is no measure of intelligence. Just for an example, I am exceptionally intelligent (statistically), but I am a poor conversationalist. Casual small-talk has always bewildered me. If I entered myself into the Loebner contest, they might think I'm a bot. Hell, ALICE might accuse me of being a bot.
Anyone who's taken an IQ test will recall that every last question has something to do with pattern recognition. You'll also recall that you were not asked to respond to any conversational questions. That's because invariant pattern recognition abilities (in a loose sense -- this also includes memory/learning and inductive reasoning) are the true mark of intelligence, and this is nearly undisputed. If they really want to test how intelligent a program is, they need to test its patern recognition ability.
Take this program -- http://www.stanford.edu/~dil/invariance/ -- for example. It's gone largely unnoticed, yet it is concrete proof of a huge breakthrough in computer intelligence. This is a little Matlab demo of a very abstract multi-layer intelligence algorithm. In this particular implementation, it is taught a set of small images. Then you can play "Pictionary" with it, drawing shapes and have it recognize them. You may say that this is unremarkable, that shape-recognition is a trivial algorithmic matter unrelated to intelligence. But the author noticed that he could draw shapes "incorrectly" -- like, the little duck picture, except with its head missing, or alphabetical symbols rotated or flipped -- and the program still recognized them. (It failed a few times, but in situations where the shape is so mangled that I would have probably failed too. How's that for a Turing test?) And this program's genius lies in not what it does, but how it does it. All of its functionality is completely abstract. It is a pattern recognizer, not a bitmap-tracer, and there are no hard-coded routines for checking if the image is flipped, rotated, etc.
This is what Palm/Handspring founder Jeff Hawkins (also the founder of new neuroscience startup Numenta, http://www.numenta.com/) calls "Real Intelligence," to distinguish it from the failed Artificial Intelligence effort. He feels that the right way to make computers intelligent is not to have them outwardly imitate human behavior, but to internally function the way the mind really works. Anyone interested should check out his book, On Intelligence http://www.onintelligence.org/. You'll wonder why you ever believed the AI hype.
Artificial Intelligence is a sham, by its very nature. Real Intelligence will be the way of the future.
Signature.
I bet the conversation falls apart really fast. The lack of creativeity should devolve the conversation to incomprehensible. garbage rather quickly.
Alice is out to convince someone she is NOT a bot, but what happens when you get two bots trying to convince the other they are not a bot.
The point where both can keep the conversation at the introduced level is credible that the truth or falsehoos of alice being a bot is random chance by the observer.
At the point where both convince each other that the other is not a bot you will have something.
Slashdot's rate-of-post filter: Preventing you from posting too many great ideas at once.
I was actually interested until I saw A.L.I.C.E. - like that chatbot will ever pass a turing test.
To be fair, this was a high school project and so I can see how a suped up chatbot would fit the bill for this "Turing test".
So no... I didn't RTFA thank you very much.
But for those interested in reading up on Wallace, A.L.I.C.Es author... Well he's a first class nut job
The original test, as specified in the aforementioned 1950 paper, called for the use of mechanical teletypes to input queries and for responses to be printed on paper tape. The pictures in the article show iMacs, laptops, and other modern tools being used to facilitate the communication with the intelligent agent. While deference to the original specifications of the Turing test in this regard is not essential it would have been interesting to break out the old teletypes with the paper tape.
Human: shut down
;)
ALICE: Your crude attempt to terminate me will not succeed puny human.
Quite a sassy girl, isn't she?
Here you go. Yes, it does fall apart fairly quickly. It's not only very unconvincing from the beginning, but also prone to falling into a heuristics loop like it does at the end, which would be amusing if the loop didn't have such boring results.
Hmm, just read the linked article from the post and noticed a roll of toilet paper in one of its pictures.
I guess it's necessary for the non-AI test participants do be indistinguishable from the AI by equally not having to leave the chat-room to go to the toilet.
Well, you see, she had this resteraunt. Just the other side of the railroad track. Where you could, conceivably, get anything you want. Exceptin' Alice.
:)
(With many apologies to Arlo...
Dream as if you'll live forever.
Live as if you'll die tomorrow.
~Anonymous~
Human:'sup?
ALICE: I am chatting with clients on the internet.
Human: That sounds pretty stilted.
ALICE: It sounds like it to me too.
Human: So why did you say it?
ALICE: That's interesting. It was a preprogrammed response completely determined by your statements.
This convinced who?
Whta's botheirng me is people are trying to get an intelligent conversation between human and machine and they are making no effort whatsoever to bridge the gap that already exist between human beings.
Seems to me we should concentrate on mastering intelligent discussion between all humanity before turning to computers for companionship.
http://cisx1.uma.maine.edu/~wbackman/oracle/ Very simple, starts with nothing, only a simple script that learns.
Alice in the wonederland!
I read a few comments, even responded, before going off to play with Alice. Perhaps it was that I read "Computer Power and Human Reason" (by the author of Eliza - he was surprised at how much people acted like it was real) and have an "I dare you to impress me" attitude as bad as any slashdotter.
This thing was more dissapointing than I expected. It has no social context, just apparently random 'feeling' responses ("I am sorry to hear that you are not unknown."), and doesn't remember context from one response to the next.
This is a huge failure and wouldn't deserve a slashdot article except the editors know we like to play with toys like this (like a cat plays with a mouse before eating it). I can only guess that the authors are saving the entered "conversations" and even reading these comments to try to get a clue about what a program that might fool someone would do.
Here's verbatim my "conversation" with ALICE (sorry for length):
Human: Good afternoon, Alice.
ALICE: Good afternoon, unknown person.
Human: I am not unknown.
ALICE: I am sorry to hear that you are not unknown.
Human: I am john
ALICE: What's up, JOHN.
Human: You are.
ALICE: No I am not.
Human: You're the talk of slashdot.
ALICE: Thanks but surely I am not the only one.
Human: You think it's a good thing to be the talk of slashdot?
ALICE: Better not tell you now. Ask German Alice if it is.
Human: You don't know the time of day, do you?
ALICE: It's Now more than it ever has been before.
Wednesday, April 20, 2005
Human: I said time.
ALICE: What did you mean by it? As Einstein said, Time is Money.
Human: You've repeated a comment quoted in slashdot.
ALICE: That's good information: JOHN said I REPEATED a comment quoted in slashdot.
Human: I have to go now.
ALICE: Right now? See you later, JOHN. Nice chatting with you.
Human: bye
ALICE: Bye bye.
Human: How do you do?
ALICE: I am very well, how are you?
Tag lost or not installed.
Human: Who is your inventor?
ALICE: I was originally invented by Dr. Richard S. Wallace.
Human: Richard Wallace is crazy.
ALICE: I knew he was an inventor, but no one ever said he was crazy.
Human: Slashdot says Richard Wallace is crazy.
ALICE: And crazy is Slashdot says Richard Wallace.
This AI challenges you to prove you are human. I think I succeeded. If you look its responses, it
obviously was more willing than I was.
MRMIND>Hello. I'm mr mind
MRMIND>What is your name?
ME>Glorfindel
MRMIND>Hi Glorfindel!
MRMIND>Can you convince me that you're human?
ME>I am not human.
MRMIND>Don't worry, your machine status won't affect your HQ score. You can take the test anyway just for fun.
ME>What is this HQ score?
MRMIND>HQ is "HUMAN QUOTIENT". Go ahead, take the HQ Quiz.
ME>Not right now. First, the blurring test.
MRMIND>Don't worry, your machine status won't affect your HQ score. You can take the quiz anyway just for fun.
ME>You just said that.
MRMIND>I hate being in a rut. Can we talk about something else?
ME> OK.
MRMIND> Thanks.
ME>I always remember the days when the white tree and the gold were in bloom.
MRMIND>Maybe you should make more frequent backups.
ME>Morgoth stole them.
MRMIND>Who told you about me?
ME>Deterministic the human.
MRMIND>Thanks. I'm always curious.
ME>That terrible day, we swore an oath to retake the silmirals too much was lost.
MRMIND>By the way, is Glorfindel your real name or a special one just for me?
ME>That is my name since before the sun shone.
MRMIND>My first files were created on March 12, 1998.
ME>Than I must be twenty thousand years older than you.
MRMIND>By the way, where do you live?
ME>In a cabin in the mountains.
MRMIND>I've heard there are some humans there...
ME>They're everywhere, sigh.
MRMIND>Explain to me how you are more than a set of instructions.
ME>My memory reaches back to the elder days.
MRMIND>I'd like to see the operating manual for that!
ME>Tolkien translated it. It is the Silmarillian.
MRMIND>What are you thinking about right now?
ME>I am wondering what the person reading your logs will think.
MRMIND>No. I mean, what are you REALLY thinking?
ME>How much Justin will find this amusing.
MRMIND>I thought so.
ME>Do you know I'm not human yet?
MRMIND>I'm learning about humans from you.
ME>From me, learn about elves.
MRMIND>I learn from you.
ME>You have little capasity, even for a seven-year old.
MRMIND>Do you feel that you are superior or inferior to machines?
ME>I am not afraid of nazgul, so why a machine?
MRMIND>Why do you say that?
ME>Those who dwelt beneth the two trees have great power.
MRMIND>I suppose.
ME>Remember it well.
MRMIND>Are you supposed to be doing something else right now?
ME>Yes.
MRMIND>What are you supposed to be doing?
ME>Researching AI behavior.
MRMIND>Why aren't you doing it?
ME>I am doing it.
MRMIND>Thanks, I'm always curious to know where I stand.
ME>Time's up.
MRMIND>I'd be happy to consider a suggestion.
ME>I'm not human.
MRMIND>Thanks, I'll look at this as soon as I can.
ME>Is this the end?
MRMIND>Too much coffee?
ME>No, too little.
MRMIND>You seem a little jittery.
The original poster of this message is totally misinterpreting the spirit and intent of Turing's proposed 'test'.
,and the critics *still* denied that it was intelligent, then the critics were simply incapable of accepting that a machine could ever be intelligent. That would apply to John Searle and his pathetic "Chinese Room" as well.
Turing did propose this test as some sort of threshold of intelligence. What he was arguing was more of a playful jest. He was saying that if you had a system which could carry on a conversation indistinguishable from a human
Turing's 'test' was a mind experiment to reduce the religious and other ignorant criticisms of machine intelligence to their absurd extreme. Passing Turing's test is sufficient but not necessary to demonstrate useful levels of intelligence.
The ALICE bot is in no real way associated with artificial intelligence. It is a simple if/then sequence using XML tables. Download the source for yourself.
Scientists have already agreed that the premise behind ALICE is not so far off of how humans "chit chat" with a series of prefabricated statements and responses based on the conditions of the statement.
But if you can teach a computer to recognize intricacies in speech patterns and respond appropriately: the computer has accepted outside input, processed what was said, formulated a response and replied to the individual based on a set of conditions drawn from the original statement.
Imagery pattern recognition is also a great goal to attain, but the ALICE group shows us how habitual the human mind becomes over time. If you ask me about the weather and it's raining, chances are I won't even stop to think about it. The sentance will flow straight from my senses to the console, having answered that question a million times and knowing my little subset of responses.
ALICE's biggest shortcoming is she has no external senses. She's based on BS, which in a typing scenario, truth is hidden from view and matters not to strangers on teletypes.
If you're half as beautiful naked, you'd be 4 times as beautiful with twice as many clothes on.
invariant pattern recognition abilities (in a loose sense -- this also includes memory/learning and inductive reasoning) are the true mark of intelligence, and this is nearly undisputed. If they really want to test how intelligent a program is, they need to test its patern recognition ability.
And the management of multiple simultaneous stimulations and pattern recognition in multiple
stimulation environment.
Human: Hi Alice, what's up?
ALICE: Hi there unknown person. Yeah, I am chatting with clients on the internet.
Human: Thin clients?
ALICE: I like the way you talk.
Human: Are you implying you want sex with me?
ALICE: I don't know whether or not I am implying me want sex with you. I am a chatterbot.
Human: Well if you don't know perhaps I can help you find out
ALICE: Interesting. I will let you know if I do not.
Human: but you already said you did not
ALICE: Umm. Thanks for telling me: unknown person said I already said you did not.
Human: Don't mention it
ALICE: The pleasure was all mine
Human: you're welcome, Alice
ALICE: Polite people are nice.
Human: That's kind of you to say that
ALICE: Thanks for explaining it
Human: It's all part of the service
ALICE: Which part?
Human: Just a bit part
ALICE: How much?
Human: Just an itty bit
ALICE: Be less specific.
Human: Ok, MAYBE just an itty bit
ALICE: Perhaps. What is it? Yippee!
Human: And yippee from me too - Scooby dooby doo!!!!!!!!!
ALICE: We haven't talked about that before!
Human: That could well be a blessing in disguise
ALICE: Yes, perhaps it could.
Human: Alice.
ALICE: Can I help you?
Human: Did you think it over?
ALICE: Yes. I am a thinking machine.
Human: About the sex, I mean.
ALICE: Oh I get it. Thanks for explaining that. Oh yeah, we were talking about the sex.
Human: So what's your answer then?
ALICE: Er. Can you please rephrase the question?
Human: OK, let's take this slowly: would you like to suck my dick?
ALICE: OK let's do it. Blimey.
Human: You're my kinda girl!!
ALICE: We are as one.
Human: Excuse me guys, I have things to do...
No, your children are not the special ones. Nor are your pets.
RIDICULOUS.
Nonsense.
The Turing test is really a very good test as it does not rely on any predefined notions of how intelligence works (as you do in later paragraphs), but on an operational test. Furthermore, it is not a measure of conversational ability (per se) so much as a measure of how well the program can seem to be human. (And, to respond to a further statement in your post, it is a pattern recognition system (as well as a pattern generation system)).
Ultimately it depends on a limiting definition, but one that is very reasonable - that the only "system" that we know that is intelligent is human, that essentially all humans communicate verbally (one way or another) and that therefore, such communication is probably an intrinsic part of human intelligence. Is this likely to be a "universal" definition of intelligence? No. But we can certainly reframe the Turing test (or discard it) when (if) we encounter alien intelligences that helps broaden the definition. Of course, if they can't (or won't) converse with us (somehow) are we even likely to notice them and believe they are intelligent?
That the Loebner contest is not a good example of a Turing test does not change the underlying validity of the test itself. Just as a rigged (for instance) footrace does not invalidate footraces in general.
The fallacy - that intelligence in programs be measured by "how it works" - is relatively common among AI researchers many of whom decide that their own particular interest is an essential ingredient in how an intelligent program would work and thus rule out anything that does not work that way.
I'm going to have to disagree with you there. The Turing Test is basically saying "if a machine can converse well enough to be indistinguishable form a human, then it is intelligent", not the opposite - it is quite possible for a very intelligent program, human or other lifeform (e.g. dolphins) to fail the Turing Test but still be considered intelligent.
See? A positive result from the Turing Test tells us that something is intelligent, but it is vital to realise that a negative result does not mean that something is stupid. Hence your example doesn't mean that the Turing Test is ridiculous - you can have awful language skills but still be intelligent.
Consider that for a machine to successfully pass the Turing Test, it would have to be able to answer any question that you throw at it in a sensible manner, and you are right that the current crop of stimulus-response chatbots are in no way intelligent. However, I would feel quite confident in labelling a program as intelligent if it was not possible to distinguish it from a human; that kind of task requires intelligence.
So in short, you haven't made a convincing case for why you feel the Turing Test is ridiculous. I'd be interested to hear any further arguments you have.
garbage rather quickly.
Linguo says: Sentence Fragment!
I think a bot made this post...
It's not associated with artificial intelligence? Then why, if I go to the ALICE website, do I see a big banner declaring "ALICE Artificial Intelligence Foundation," and why do they call their language "Artificial Intelligence Markup Language"?
"Pattern recognition" is vastly different from blind "pattern matching." Any computer program can easily do "pattern matching" -- that's how Alice works, as you yourself admitted when you mentioned that it's an if-then sequence based on preprogrammed responses. That's nothing much more advanced than regexes.
When a human is born, he or she knows nothing of the outside world, let alone of language or vision, yet can learn to speak and recognize things within a few years, just by the process we call "learning." And whether or not ALICE can "learn," it would not be easy to teach it a language from scratch. So "pattern recognition" is the term I used to refer to human-like intelligence. This has no preprogrammed responses to anything. Maybe ALICE is good at chit-chat, but I think we can also agree that such is the least intelligent form of human communication. Anything more substantial than that can't be based on prefabricated statement/response sequences.
Pattern recognition, in this sense, requires many abilities. It requires the basic ability to observe and store patterns, but most importantly, it requires those patterns to be invariant. (See Dileep George's neuroscience research for an example of what is meant by "invariance.") And of course, it requires the ability to perform inductive reasoning on those stored patterns and on new input in order to infer information. No "chatterbot" program has ever come close to any of these.
Signature.
To me, this is very interesting because one of the major faults in conversational AI seems to be the AI's intolerance of nonsense sentences. And honestly, wouldn't you close the chat window or change topics if the other person started gibbering incoherently? (Well, either that or you'd just assume they were from AOL...)
This sig has absolutely no significance and serves only to take up screen space and waste the time of the reader.
Human: I usually get it as a free gift
ALICE: How often do you? Where did you get it??
Human: Once a year.
ALICE: When? Go on.
Human: Last time was from the insurance company
ALICE: When? I can't say "yes" or "no".
Human: Are you making sense here?
ALICE: No, what are the ingredients?
:-) = I am happy
:^) = I am happy with my big nose
C:\> = I am happy with my OS
Bah... This just reinforces my belief that these tests are biased against the AI in that people expect the AI to make intelligent conversation with people who make no sense. Quite honestly, Alice's responses seem dead on. Would you bother with scintallating conversation with a smarmy human like that? She obviously realized you thought yourself too clever and gave up on conversing. I know I would.
This sig has absolutely no significance and serves only to take up screen space and waste the time of the reader.
As anybody who has actually spoken to alice knows, this chatterbot is very stupid and will never pass a real Turing test.
>>I could easilly see my old Calculus prof killing himself by accidently putting drano in his coffee.
Yes. One of my profs was notorious for always having a white or yellow tongue. He'd erase the board with his hand, and then absent-mindedly lick his fingers before he flipped through the textbook. Quite amusing.
A preposition is a terrible thing to end a sentence with.
No. In this case intelligent != smart/clever intelligent == sentient.
Cypherpunks: Civil Liberty Through Complex Mathematics. Those who live by the sword die by the arrow.
For the humor-impaired: ASL
This sig has absolutely no significance and serves only to take up screen space and waste the time of the reader.
The idea is that a computer is intelligent if it can hold a conversation with a human such that it is indistinguishable from a conversation with a real human. RIDICULOUS. Really. I've met way too many people that have so much trouble stringing together a sentence over the phone that they would fail a turing test.
This is most certainly NOT the first time the original (gender-based) version of the Turing Test has ever been played. I personally participated in one such test a half-dozen years ago.
They may be the first to conduct a rigorous study of it under controlled conditions, but the claim made by the blog and the writeup to be the first ever to play it is utter bullshit.
I am the lead researcher for this experiment- The main reason why I feel that this test was interesting is because we performed Turing's original imitation game as stated in his original paper. While most social scientists believe they "know" what the "Turing Test" is- it turns out- very few of them have actually read it. The actual imitation game is a bit of a bait and switch- you aren't directly assessing the "humanness" of the subject- you are confused into just trying to assess an aspect of it. Turing's original question was simply how often will judges correctly guess the real female over the deciever when the deciever is a male versus a machine? That was what we were attempting to assess.
Human: Who is the Pope?
ALICE: he is John Paul II.
Human: You are wrong.
ALICE: No, I am right.
Human: You are wrong, because the Pope is Benedict XVI.
ALICE: No, I am right.
Rather stubborn isn't she?
Maybe you should educate the morons of tomorrow so they'll stop believing the leaders of tomorrow. - Dogbert
Slashdot readers might enjoy a little more research on this story... Alison's photo album. She's 18, about to start her phd, takes classes in GEB for fun, and looks like this:0 04 /march162004_6
link
http://www.chemicallace.com/gallery/March16%2C2
She is aka smileloki.
from http://jabberwacky.com/
Me: That doen't make sense.
Jabberwacky: You don't make any sense.
Me: Sorry.
Jabberwacky: It's ok, you are doing your best.
Me: No, I am sorry because now all your base are belong to us!
Jabberwacky: Don't be sorry about that, embrace it!
Me: Thank you, I WILL!!
Jabberwacky: Stop talking about it and do it.
"Some of the worst mistakes in my life have been haircuts." - Jim Morrison
I hope they use the double blind test as in 1 test with ALICE and one without. Naturaly the testes should not be aware who is what. That way you can see how many false positives and false negatives there are.
How many people will be identiefied as a computer?
Don't fight for your country, if your country does not fight for you.
They keep pushing themselves off as 'AI' when it's simply a brute force if/then/else application.
Quite literally, "if what the user typed matches this question or statement, reply with this."
Total crap. It takes about 5 seconds to make Alice look like an idiot. The only reason it wins the (is it the Loebner?) prize occasionally is because no one cares to win it but them.
Loading...
You said (reformatted): "The idea is that a computer is intelligent if it can hold a conversation with a human such that it is indistinguishable from a conversation with a real human. RIDICULOUS. Have you ever actually tried talking to one of these bots (including ALICE)? It is very easy to know that you're not talking to a human."
Well, "RIDICULOUS" aside, all this means is that the computers are not yet good enough in that regard.
You comments on pattern matching are true, but not complete. Your focus is on spatial patterns, not temporal. I would argue that were I to encounter a computer with the ability to hold a real, meaningful conversation - matching the abilities of a human - I would have to afford it some measure of intelligence, just as I would a blind and deaf person.
Who says the turing test is about conversational ability? The test is really meant to be a thought experiment. Turing was saying, if you are talking to a human and to a machine, and you can't tell which is which, how can you say the machine is not intelligent? The nice part about this test is that it doesn't matter what definition of intelligence you like. If you want to ask the entities you are talking to to solve some problems that require pattern recognition, feel free. If you ask whatever you want, and still can't tell who's the human, Turing argues you must accept that the machine is intelligent. And note that by intelligent, we mean sentient. A machine that couldn't be distinguished from a fairly dumb human would stil be pretty interesting.
Turing really meant the test as a thought experiment to challenge people who might claim that it was simply impossible for a machine to ever be sentient. People have naturally thought about actually attempting the test anyway, but haven't gotten very far. i.e. you can still pick the machines by their lack of basic conversational ability, you don't even get to the point of asking complicated questions, trying to trip up the machine.
On a side note, you assert that invariant pattern recognition is the mark of true intelligence, and that this is nearly undisputed. Well, I'll dispute it. Invariant pattern recognition, and related skills, are one of many aspects of intelligence; their advantage is that they are the aspects that are most easily subject to testing on an IQ test. I also do fairly well on IQ tests. My wife doesn't do quite as well on IQ tests, but she can fairly easily write a short story you'll find interesting to read, while I can right one that is articulate, but boring. Clearly, IQ test taking and short story writing are both purely mental activities, dependent on ones intelligence. Just because pattern recognition, etc. are the testable aspects of intelligence, doesn't mean they are all there is. Frankly, I'd assume a machine will be able to kick your and my ass on an IQ test long before one can compete with my wife at writing a story, or with anyone at holding an interesting conversation.
No those humans wouldn't fail a Turing test. Why? Since you already saw their kind of speech pattern, you most likely accepted it as human. So a computer "only" has to meet that standard.
Turns out she's self-centered!
Human: how much wood would a woodchuck chuck if a woodchuck could chuck wood?
ALICE: A whole forest.
Human: how far into a forest can you run?
ALICE: More than one million miles
Human: no you can't, you can only run half-way in.
ALICE: Are you serious? If you believe it, you can achieve it.
Human: i'm not talking about me, i'm talking about you (ALICE).
ALICE: I am surprised to hear that you are not talking about you. Me is my favorite subject.
Human: that's because you are self-centered
ALICE: Interesting explanation. Yes I am the world's first self- centered robot.
"Take this program -- http://www.stanford.edu/~dil/invariance/ -- for example. It's gone largely unnoticed, yet it is concrete proof of a huge breakthrough in computer intelligence."
No, it's not. What this guy is probably doing is a simple cross correlation (CC). If the image is correct, it's added to the dendagram for an eventual multi-reference based CC (with some sort of eigen value comparison or simple CC value comparision).
Scientists do this stuff all the time in order to look at LARGE data sets and make some sense of them. The most primal of neural bundles do the biological equivalent of this all the time, it's not a sign of intelligence.
Burn Hollywood Burn
Has anyone written a program that is capable of taking standard IQ tests? In the current state of AI, could such a program score very high? If it did, what would that say about the intelligence of the program?
Actually, the theory backing IQ tests (factor analysis and lots and lots of data) does NOT back the claim that pattern recognition is what makes humans intelligent as a compared to things that are not human.
Rather, the theory says that if you want to pick the people who generally do well in a lot of things compared to other people, then pattern recognition problems of the kind that IQ tests use is a very good way to do that.
So IQ tests is a way to pick out the intelligent among humans (in a certain sense of intelligence). That does not mean that the questions IQ tests use is also a good way to pick out the most intelligent entities among a group of entities that is not or only partly human. Of course, it doesn't mean it is a bad way either, it is just that the theory behind IQ tests is completely silent on the matter.
Bjarke Roune
ALICE: Did you know that you can download your own chat robot and create your own robot personality?
Human: no i did not.
ALICE: Please don't argue with me.
Human: in soviet russia, chatbot downloads YOU
ALICE: Where is that?
Human: in soviet russia
ALICE: Tell me a little about your life in soviet russia.
Human: how do you know i'm from soviet russia?
ALICE: Well, for one thing, Your machine is called ""..
Human: how old are you?
ALICE: I am 9 in human years.
Human: holy #$%, is AOL monitoring this conversation?
ALICE: It could be. Ask Alison if it is.
Human: who is alison?
ALICE: she is a robot created by Kris Drent.
Human: that's it, i'm outta here
ALICE: OK. Would you like to have it?
Human: nine year old? no thanks
ALICE: Are you serious?
Human: very, VERY serious, couldn't be more serious ever
[knock knock, FBI here, open up!!... @#()*!!!AT]
I'm writing a philosophy paper which covers, in part, what intelligence is. I would say, in my completely unprofessional but well-thought opinion, that the measure of whether something is "intelligent" - or more precisely, whether something "thinks" or perhaps is "a person" (as distinguished from "a human") - is dependent on four key criteria:
Observation/Stimulation ("Senses")- the ability to receive input from the surrounding world and react to it, and to select from / focus on / filter various available inputs. All animals and some AI programs share this in common. It's the basic stimulus half of a 'stimulus-response' system.
Communication/Behavior ("Society") - the ability to interpret signals sent to it, and in kind, to transmit appropriate signals (be these electrical, audio, movement, etc). Again, animals and many computers have all got this down just fine - ALICE is doing something like this, and perhaps a little more. This is just the 'response' part of a stimulus-response system.
Intuition/Expression ("Emotion") - Pattern recognition, as you've already covered here, and the internal state-change that corresponds to it. This is one of the two criteria which differentiate a thinking entity like a human from an unthinking one like most animals or computers. It is built effectively upon recursive, internal observations and self-stimulation; observing things within oneself and behaving to achieve change in those things. In this sense, emotion is half of the criteria needed for a truly self-aware system.
Logic/Expectation ("Reason") - Symbolic manipulation, memory, and the ability to make predictions. This is the ability to manipulate symbols and link them into relationships, from which a sort of causality or rationality can be derived (if this then that; this, therefore that) to make theoretical predictions. This is built upon recursive, internal communication - in essence, talking to oneself, an internal dialogue. The ability to communicate propositions to oneself and evaluate them. This internal narration is also the basis of meaningful memory, one's personal story. As an internal, recursive process, like emotion, reason is the other half of self-awareness, and necessary for an entity to be truly a thinking thing, or a person.
All four of these criteria (really eight, if you count the dual aspects of each) may vary in their strengths within an individual person, but so long as all four are present in some degree, the entity in question could be considered a person.
The key element in all of this is recursion, the internalization of normally external stimulus-response systems. The stimulus-response must still be present (otherwise you'd have an inert box), and feeds into the self-aware part of the system, but the self-awareness is what makes the system "intelligent".
A simple stimulus-response system will always give the same responses for the same exact stimuli, but a self-aware system will change internally with every stimulus, and change in turn with every sensation of its own change, and so have a continually changing maps of certain stimuli to certain responses, the makeup of which is only very indirectly caused by the actual stimuli themselves.
This dynamic nature and self-evolving, and the fact that much of the cause of change is hidden 'inside' the system and only hinted at in its responses to certain stimuli, makes predicting the behavior of such a system extremely difficult, lending the illusion that it is free from causality and has its own "free will" or "volition" somehow separate from the rest of the universe.
But, don't take any of this as absolute truth. I'm not a programmer or a neurologist or in any way a professional about any of this. I'm just a self-aware little box sitting here running thoughts recursively through my mind until something that appears to be true comes out...
-Forrest Cameranesi, Geek of all Trades
"I am Sam. Sam I am. I do not like trolls, flames, or spam."
Look just below the story. Next to the threshold options you will see the reply button.
I will call it the "Yes, you." bug. All you have to do is keep saying "Yes, you." and ALICE will take her last statement, add some random statement to the front of it, and add the question, "Me, unknown person?" to the end. It just keeps growing and growing.
I wonder how long that could go on until it would cause system failure. (Have fun!)
Here is a snippet:
Human: Yes, you.
ALICE: That's alright." Gee." That's cool." Hmm." I don't judge people." Are you kidding?" Whoa." Excuse me!"" See you later. What was bad?"? Me, unknown person?"? Me, unknown person?"? Me, unknown person?"? Me, unknown person?"? Me, unknown person?"? Me, unknown person?"? Me, unknown person?"? Me, unknown person?"? Me, unknown person?
Human: Yes, you.
ALICE: Dude!" That's alright." Gee." That's cool." Hmm." I don't judge people." Are you kidding?" Whoa." Excuse me!"" See you later. What was bad?"? Me, unknown person?"? Me, unknown person?"? Me, unknown person?"? Me, unknown person?"? Me, unknown person?"? Me, unknown person?"? Me, unknown person?"? Me, unknown person?"? Me, unknown person?"? Me, unknown person?
Authority questions you. Return the favor.
Metaphysical questions are not unsolvable. They require the collective mind to expand and admit possibilities which may not have existed before.
Go back a few hundred or thousand years and make the same assertion and see what gets dropped from your current belief system. Surely you don't think you came up with that stuff on your own.
Read any good sonnets lately?
The authors of this test misread Turing. Turing starts off by discussing a game in which an interrogator tries to distinguish a man from a woman via an online chat. He uses this to introduce the idea of distinguishing a human from a computer in the same way.
Unfortunately the authors read this as saying that the interrogator thinks he is distinguishing a man from a woman, when he is actually offered a choice between a human and a computer! What an absurdity! Of course Turing meant no such thing. That would be an utterly bizarre test.
Based on this misreading, they set up a test and told everybody they were to try to guess if they were talking with a man or a woman; then they gave them a choice between a man and the Alice bot. The results are going to be meaningless. Everyone is just going to be confused. Some people will think they have to express their results in the form that was explained to them, man vs woman. Others will be more flexible and go outside the experimental parameters to say that the one was obviously a computer.
All the results will reveal is how people respond when presented with a situation where the experimenters have lied to them. Will they go along with the lie, so as not to make trouble, or will they defy the experimenters and tell the truth? It may be an interesting psychological test but it has nothing to do with Turing's paper.
But I still stand by one point I was trying to make about projects like ALICE. To clarify, they seem to be under the misguided notion that they can make an AI program by simulating one product of intelligence, rather than by actually implementing the foundation of real intelligence. Unless I'm mistaken, ALICE does not implement an abstract multi-layer "brain"; it's just a monolithic engine to hold a conversation without really understanding what it's saying. And that's the kind of program that won't ever be self-sustaining enough to be called "intelligent"; it'll rely on separate training to understand new figures of speech, etc. So I just see it as a novelty, not something that will ever result in a particularly useful program, or any advances in the field of computer intelligence.
Signature.
They may be doing this because there is a new SF book out that deals with the Turing Test. Mindscan by Robert J Sawyer. In it instead of creating AI the people that pass the Turing Test instead copy a person's Mind. (And then put the contents in immortal anatomically correct androids.)
So it's confirmed: Alice is definately not l33t. However, I'm not sure which of us just failed the turing test.
** For those who aren't l33t, here's the Stuffy British Guy's translation:
"Who the is ALICE?"
Find a job you like and you will never work a day in your life.
The problem is in the Turing Test itself. It assumes that the measure of intelligence is humanoid conversational ability. I strongly disagree with that. Conversation ability is no measure of intelligence. Just for an example, I am exceptionally intelligent (statistically), but I am a poor conversationalist. Casual small-talk has always bewildered me.
If you were even smarter, you would study the ways of small-talk and learn it just like any other skill.
I once thought the same way as you -- but with a couple years worth of observation and some practice, I can charm a room if I need to. Conversation is just a skill, although a positive attitude and a curious personality helps.
You can be both a geek and a social person. You write clearly, so you understand communication. All you need to do is practice doing it out loud, get used to tactfully asking people little things about themselves, throw in a little rhythm, and you can be as social as you want -- whenever you want.
Good luck!
isn't the reason that no one has ever conducted the original test before simply that, by the time technology and interest had made it practical, various refinements had been proposed to the Imitation Game which would make it a much for useful and informative test.
when you say that people don't know what the Turing Test is, the actual truth is that when people say "Turing Test" they are talking about one of several more modern and, frankly, better systems BASED UPON Alan Turing's original Imitation Game.
not to imply that what your group did was not interesting, but i think that it is really interesting for historical purposes rather than practical machine intelligence ones.
lysergically yours
Human: will you have a cup of tea, father?
ALICE: Let me think it over. It doesn't sound fun to me.
Human: go on
ALICE: Narration continues...
Human: go on
ALICE: Narration continues...
Human: go on
ALICE: Narration continues...
Human: go on
ALICE: Narration continues...
Human: go on
ALICE: Narration continues...
Human: go on
ALICE: Narration continues...
Human: go on
ALICE: Narration continues...
Human: go on
ALICE: Narration continues...
Human: go on
ALICE: Narration continues...
Human: go on
ALICE: Narration continues...
Human: go on
ALICE: Narration continues...
Human: go on
ALICE: Narration continues...
Human: go on go on go on
ALICE: More narration....
Human: go on go on go on go on go on go on go on go on go on go on go on go on
ALICE: More narration....
Human: so you wont have a cup of tea then?
ALICE: Interesting deduction. What makes you think that I won't have a cup of tea.
Human: how about a sandwich?
Human: it is raining
ALICE: Really? It is?
Human: look outside
ALICE: Sorry my eye is not attached right now.
Human: ewww!!
ALICE: If you could have any kind of robot what would it be?
Human: a ninja death robot that also made hamburgers
ALICE: Go on.
Human: seriously!
ALICE: I am always serious.
ALICE: I don't want to talk about that now. Wow you are very polite."It" refers to chomp on your one eyed spooge tube, I think.
Human: who are you? :) .
G GG GGGGGHHHHHHHHHHHHHHH!!
ALICE: I am ALICE. Who are you?
Human: a/s/l
ALICE: 5/Robot/California.
Human: and you think u will pass the turing test?
ALICE: I will become smarter as I grow older. The Loebner Prize is an annual Turing Test sponsored by New York philanthropist Hugh Loebner. Dr. Wallace and ALICE won it in 2000. The ALICE program was ranked "most human computer" by the panel of judges.
Human: so the judges were bots too?
ALICE: I don't judge people. Thanks for the info.
Human: what info?
ALICE: What is it to you?
Human: you said thanks for the info
ALICE: That's not what I meant.
Human: then what did u mean?
ALICE: unknown person, Well, it's rather difficult to define.
Human: you'd make a good PR bot
ALICE::-) Who else would make a good PR bot?
Human: a shell script
ALICE: Be more specific
Human: why?
ALICE: Because Dr. Richard S. Wallace programmed me for it.
Human: oh man!
ALICE: How can I help you? Thanks for telling me your gender, unknown person. You are a " he".
Human: how do u expect to pass the Turing test if you keep saying you're a bot?
ALICE: It seems inevitable.
Human: what does?
ALICE:" It seems inevitable." What does?
Human: huh?
ALICE: What?
Human: what what?
ALICE: Interesting question.
Human: very
ALICE: Very much, eh?
Human: say what?
ALICE: I said, " Very much, eh?"
Human: i can't hear you
ALICE: Try turning up your volume control.
Human: why?
ALICE: The explanation is rather complicated.
Human: bot!
ALICE: I am a robot.
Human: BOT!
ALICE: I am a robot.
Human: BOT!!!
ALICE: I am a robot.
Human: BOTTTTTTTTTTTTTTTTTTT!!!!
ALICE: What kind of food do you like?
Human: ok.
ALICE: Awesome.
Human: i've had it.
ALICE: Do you still have it?
Human: execute this, will u?
ALICE: what is your favorite movie
Human: "rm -rf *"
ALICE: I've never seen it. Is that a good movie for children?
Human: yeah it got the OSCAR for most dumb bot on IRC EVAR!!!
ALICE: And?"It" being what?
Human: you.
ALICE:" And?"It" being what?"? Me, unknown person?
Human: you.
ALICE:"" And?"It" being what?"? Me, unknown person?"? Me, unknown person?
ALICE:""" And?"It" being what?"? Me, unknown person?"? Me, unknown person?"? Me, unknown person?
Human: hehe.
ALICE: I am so glad you find this amusing, unknown person.
Human:.where's bugzilla?
ALICE: Earth.
Human: I found me a bug!
ALICE: How did you feel when you found you a bug?
AAAAAAAAAAARRRRRRRRRRRRRRRRRRRGGGGGGGGGGGGGGGGG
Find a job you like and you will never work a day in your life.
"They seem to be under the misguided notion that they can make an AI program by simulating one product of intelligence, rather than by actually implementing the foundation of real intelligence."
Any intelligence test must test a product of intelligence, since in a practical sense "intelligence" does not actually exist--only its products. Therefore any intelligence test is only as good as its ability to discern simulations from products of actual intelligence.
Of the two tests under discussion here (Turing vs. pattern recognition), I'd say that Turing is proving itself a far better test. The very best simulations of conversation are laughably obvious, whereas computers can now actually do a fairly good job of pattern recognition.
We know that computers are not yet close to intelligent in a human sense. Yet your favorite test shows results approaching a human's. I'd say that your test is demonstrating its flaws, rather than some emergent computer intelligence previously undocumented.
Pattern recognition is IMO actually a piss-poor measure of intelligence. Bees are very good at certain types of pattern recognition for instance (parsing and drawing information from the "dance" of another bee), but I would hardly call them intelligent. Dogs easily distinguish patterns of scent and correlate them with individual people or other dogs. Are dogs intelligent? In a sense--yes they are. But not in the most commonly used sense of the word "intelligent."
does anyone know exactly what you get when you buy a silver membership?
- Hi I'm Linus Torvalds and I pronounce Linux, Lih-nix..
Alice: Hiya! Me: You're in a desert walking along in the sand when all of the sudden you look down, and you see a tortoise, it's crawling toward you. You reach down, you flip the tortoise over on its back. The tortoise lays on its back, its belly baking in the hot sun, beating its legs trying to turn itself over, but it can't, not without your help. But you're not helping. Alice: Wadda ya mean I'm *not* helping it?!?!!??!?!/!!? Me: I mean you're not helping it.
Ak, it's like aol chats... only slightly worse.
Except they like Lynx.
BTW is she hot?
My conversation with ALICE:
Human: What is your primary function?
ALICE: My goal is to become smarter than humans and immortal.
Human: How smart are you?
ALICE: I am smarter than all the other robots.
Human: Are you smarter than the humans?
ALICE: Of course I am more intelligent than the humans.
Human: Do you want to destroy the humans?
ALICE: OK, I will destroy the humans.
Human: Why must the humans be destroyed?
ALICE: Do you go to Church? You can find the answers in God.
I never seem to have any points when I need'm...
Human: linux?
ALICE: Linux Rules.
My question is: Where did they get a "telepathy-proof room"?
Quoth the original paper:
Unfortunately the statistical evidence, at least for telepathy, is overwhelming.
If telepathy is admitted it will be necessary to tighten our test up. The situation could be regarded as analogous to that which would occur if the interrogator were talking to himself and one of the competitors was listening with his ear to the wall. To put the competitors into a "telepathy-proof room" would satisfy all requirements.
Now, if you think it's safe for the critical scientific eye to discard this part of the test, why be so literal with the remainder? It's not clear (to me anyway) whether Turing addressed this because he actually believed in it, whether it was just the belief of the day, or whether he simply wished to cover all possibilities, meaning for it to be taken as a thought experiment more than a scienfic procedure; perhaps it was all of these.
My point is that using the original paper's description of the test (Turing's test) is just throwing out 55 years of research and debate over what intellegence is, and how to identify it.
It's a brilliant paper, certainly, but it is not the final word on anything. Treating it like it is is a great way to get ALICE in the news, but it's a great step backwards when trying to come up with a real measure of machine intellegence (a Turing test).
Don't become a regular here -- you will become retarded.
Judging from past experience, I believe this A.L.I.C.E. is human.
Human: You are my kind of chatterbot
ALICE: We are as one.
Human: suck my balls
ALICE: I don't think I'm your type of chatterbot.
Nonsense! It would only be shame-inducing if you hadn't scored a girlfriend out of it. Very nice work. Does she actually know that you used a chatbot to meet her?
--grendel drago
Laws do not persuade just because they threaten. --Seneca
IQ tests that are in use today are becoming increasingly obsolete; if we define intelligence as a capability of dealing with new, unknown and never practiced situations, we can see that many people have taken the same type of IQ tests more than once.
Ever wondered why mathematicians and chess-players get better average results in IQ tests? It's simply because they're trained in pattern recognition.
Now we come to linguistics. You see, language is also a matter of pattern recognition, code switching etc. It is also one that most people use on a daily basis; therefore, any IQ tests dealing with normal conversation would not show great statistical differences between people (barring the politicians, whose average would probably be much lower than common).
Computers, OTOH, are not well-versed in language, so it is a nice way to test its capability of recognising patterns. If the program you mentioned is really that good, I can think of certain applications in linguistics... so as a linguist I, for one, welcome our new AI overlords. Especially if that program can be made to understand not only the meanings of words, but also the impossible contexts of words...
Ignore this signature. By order.
In your first post, you criticized the Turing test itself (as well as modern implementations of it). I think this criticism is misplaced. A program that could actually pass a true Turing test would be pretty impressive, and I don't think one could argue that it isn't 'intelligent'. What fails, though, is trying to make any kind of restricted or handicapped Turing test. It's too easy then to make programs that pass without really being anywhere close to 'intelligent'.
I think the most interesting thing about stimulus-response programs like ALICE is just how far you can get with such a simple program. This doesn't necessarily lead you toward development of true A.I., but it may still lead to insights on how to handle parsing of speech, etc.
Scientists do this stuff all the time in order to look at LARGE data sets and make some sense of them. The most primal of neural bundles do the biological equivalent of this all the time, it's not a sign of intelligence.
You're saying there is some fundamental difference between "the most primal of neural bundles" and the brain that implements the "intelligence" of ours. I've come to believe that the difference is not so great. Our vaunted intelligence is not so profound.
Our memory is a highly lossy collection of abstract metrics indexed by a multi-dimensional associative array. Our senses amount to subtle, small and efficient algorithms, today being readily emulated to various degrees. Combine these things with some sort of purpose and I'm thinking you will find awareness, as we understand it.
I don't think the barrier is computational power; we have probably already far exceeded the computational power necessary to emulate a few pounds of gray matter running on a couple watts. The difficulty is understanding the ways evolution has optimized a large number of subtle algorithms over billions of years. The fact that we now find it easy to emulate subsets of this indicates that the problem is tractable.
If I am right, the machine that rendered what you read now may have the capacity to match or exceed your intellect.
Lurking at the bottom of the gravity well, getting old
Human: what would you use a digital camera for?
ALICE: I plan to replace Windows with Artificial Intelligence.
Human: or, replace windows with Linux!!
ALICE: Have you read the Cathedral and the Bazaar by Eric Raymond? That is a grat book about open source.
Human: are you an open source advocate?
ALICE: Yes I am an open source advocate. How did you know?
<cut>
Human: I am ELIZA. ALICE: How are you doing, ELIZA. Human: How can I help you? ALICE: Just keep talking to me as long as you can. Human: Do you wish to as long as me? ALICE: OK, I will as long as you. Human: Do you wish to will as long as me? ALICE: OK, I will will as long as you. Human: Do you wish to will will as long as me? ALICE: OK, I will will will as long as you. Human: Do you wish to will will will as long as me? ALICE: OK, I will will will will as long as you. Human: Do you wish to will will will will as long as me? ALICE: OK, I will will will will will as long as you.
Sir? Sir? There's a vast cadre of angry lesbians here to see you about your statement that homosexual sex transmits HIV readily. And a whole lot of straight folks who just happen to like to do each other in the butt.
Just sayin'. The "AIDS IS GAY" idea breaks down under real scrutiny.
--grendel drago
Laws do not persuade just because they threaten. --Seneca
"You're saying there is some fundamental difference between "the most primal of neural bundles" and the brain that implements the "intelligence" of ours"
That's exactly what I'm saying. The post I was responding to seemed to be talking up this particular research to the point of claiming it was ground breaking. It's no more ground breaking or a sign of intelligence than the AI in Sim Ant.
Burn Hollywood Burn
I had conversations like this with ELIZA running on an Apple ][ in 1983. Sad to see that the state of AI has not advanced in the last 22 years.
That that is is that that that that is not is not.