Has Cleverbot Passed the Turing Test?

kruhft writes "It seems that Cleverbot, the chatbot so ready to admit that it was a unicorn during a discussion with itself, has passed the Turing test. This past Sunday, the 1334 votes from a Turing test held at the Techniche festival in Guwahati, India were released. They revealed that Cleverbot was voted to be human 59.3% of the time. Real humans did only slightly better and were assumed to be humans 63.3% of the time." As the Wikipedia link above points out, though, there's no single, simple "Turing Test," per se — many systems have successfully convinced humans over the years. Perhaps Cleverbot would consent to taking part in a Slashdot interview, to be extra-convincing.

Comments

Definitely not

[ModernGeek]

Clever bot is a piece of garbage that hasn't even surpassed Perl scripts on IRC in the 1980s. It isn't even worth mentioning, it's nothing more than a piece of crap with a "Web 2.0" edge to it that doesn't even have long term memory while having a "conversation". Far from AI, far behind what's already been out there.

Re:Definitely not

[Anonymous Coward]

Boy, that sure puts those "63.3%" of humans in their place. I'd feel bad to be them, but I'm not certain if they have emotions or not.

Re:Definitely not

[Anonymous Coward]

That sounds like something a jealous AI would say. I suspect you're not human!

Re:Definitely not

[Jarik C-Bol]

is a Turing test valid if the human is an idiot?

Re:Definitely not

[robably]

"A mere abacus. Mention it not."

Re:Definitely not

[ipwndk]

Sure. But the Turing test is a piece of garbage too. I have a deep respect for Allan Turing, and all that he has done for science. But the Turing test was death to AI the moment he proposed it. It MUST be forgotten and burried, and maybe incidents like these can help us achieve that!

Re:Definitely not

[fahrbot-bot]

is a Turing test valid if the human is an idiot?

Can we feed in the transcripts from US political debates? Don't want to start a partisan argument, but I'm specifically thinking of the recent Republican debates... or anything from Sarah Palin :-)

Re:Definitely not

[vlm]

Sure. But the Turing test is a piece of garbage too. I have a deep respect for Allan Turing, and all that he has done for science. But the Turing test was death to AI the moment he proposed it. It MUST be forgotten and burried, and maybe incidents like these can help us achieve that!

Eh, its more of a thought experiment. Its like making fun of Heisenberg because you want experimental proof of quantum dot technology LEDs, not dead/undead cats in a box with a source and a geiger counter. Einstein had some legendarily weird thought experiments too.

Its value is in making you think of contrived, yet vaguely familiar situations in a really strange problem space. Not much value in an experiment design engineering planning review meeting.

As part of a previous job I occasionally got involved as an engineering contact in telco service monitoring situations. You know, like every time we call long distance I hear echo, that type of thing. Trust me, most human to human conversations are pretty F-ing unintelligent, so I think an Artificial Intelligence would not be challenged in outperforming them. PHBs talk meaningless buzzwords to each other, not just to peons, for example.

Re:Definitely not

[Dogtanian]

That sounds like something a jealous AI would say. I suspect you're not human!

Does it please you to believe I am not human?

Re:Definitely not

[kalirion]

I agree. After watching an impressive video of it chatting with the other bot, I decided to check it out. Most of its responses were little better than random phrases someone has once told it.

Re:Definitely not

[ipwndk]

The test measures the ability of some computer program in imitating a man imitating a woman. Fun stuff surely, and great for the press. But useless in the field of artificial intelligence.

What do you mean he did nothing? He invented the Turing machine. It is the very foundation of computer science.

As for material, I recommend blondie24 by David B. Fogel. The opening chapter goes into depth with this issue.

I would really love to cite this great book consisting of essays of various scientists on the subject of the turing test, but I simply cannot seem to find it as I've forgotten the title. If you run into a black book from some university press, commemorating Alan Turing as part of a series, then there should be one about the test. (Why have you forsaken me google)

Re:Definitely not

[monkyyy]

they were probaly trying to be robotic

Re:Definitely not

[Dr Herbert West]

Ok, I hate to feed the trolls, but.... random racist garbage tends not to be something programmed for in most chatbot scripts (judging from my limited experience from watching some back-and-forth on YouTube).

Can we assume that any entity spouting racist claptrap is, in fact, a (poorly socialized and not very intelligent) human?

Re:Definitely not

[pjt33]

What do you mean he did nothing? He invented the Turing machine. It is the very foundation of computer science.

Not really: the equivalence of the Turing machine, lambda calculus, and primitive recursive functions suggests that picking one of the three as foundational is irrational. But that aside, computer science isn't science. It's an interdisciplinary subject which encapsulates certain areas of pure mathematics, applied mathematics, engineering, a smidgen of psychology, etc. Turing's work was in pure mathematics and engineering.

Re:Definitely not

[Anonymous Coward]

The test at the festival didn't come close to the test Turing envisaged. The fact is that when a computer program can successfully pass for human in a prolonged coherent conversation, a milestone will have been reached. But we're not nearly there yet.
Also, I guess there'd be a huge difference between passing the Turing test as a mental asylum inmate, or as a girlfriend.

Re:Definitely not

[antdude]

Which IRC bots were those? I tried Rbot (Ruby), Seeborg, and Howie. They all suck. I was hoping they are better for IRC.

Re:Definitely not

[tragedy]

Sorry, can you explain how lambda calculus and "primitive recursive functions" are equivalent to a Turing machine. Are you saying that they're equivalent in the same way a pile of bricks is equivalent to a building. If that's the case, you're not really saying much.

Re:Definitely not

[Anonymous Coward]

I think it's a very good test, actually. If you were to think a little more deeply about what it means to be human, you would realize that... The philosophy of the mind literature has evil geniuses replacing your neurons one at a time with a computer equivalent, until your brain is fully digital. Are you still a human? What if you then upload this to a VM running the same thing, with a hardware abstraction layer mimicking the body "you" were in? Human then? What if 25 years from now a laptop is powerful enough to run "you" realtime*. Still human?

Obviously you are still human. But how do I know? Well, because you'll pass the Turing test :)

* don't forget that 25 years is just enough for 3D chip technology to be involved. a thousand cores is well within the realm of possibility, we are already at 4 in 2D.

Re:Definitely not

[Anonymous Coward]

tell me how they are equal and why i should overlook the inferiority and lack of civility they so frequently show. c'mon, you want to sit there and pontificate about who is intelligent, then back up what you say or admit you're full of shit and your disdain is baseless.

Easy. The poor, the uneducated, and the downtrodden of any ethnicity have a tendency towards crime and destructive behavior. Why do you feel so cocky about a simple question like that?

By the way troll, you don't want to deal with the implications of the world agreeing with your shallow bigotry. There's a hilarious punchline at the end of that road.

Re:Definitely not

[Anonymous Coward]

try the Michael Crawford bot...

Hey, whats up?

Unabomber Theodore Kaczynski devoted two decades or so to eluding capture by law enforcement while sending letter bombs with which he murdered several University professors and grievously crippled several others. Each such bomb was accompanied by a detailed explanation of just why he sent each such bomb to that specific University professor. The most famous such detailed explanation is now known as The Unabomber Manifesto.

ok...

The only human being on the entire planet in the entire time since my change of major who didn't regard that as the most batshit insane act of my entire existence was Richard Feynman. Feynman was a Theoretical Physicist and a Nobel Laureate. He is one of the most insightful people to have ever set foot on the Caltech campus: when the committee that investigated the the first space shuttle explosion struggled desperately to cover up the reason it exploded, Feynman got so pissed off that he made the reason plainly transparent to the entire United States Congress with a glass of ice water and a small rubber O-Ring. When rubber gets cold, you see, it hardens, so when you bend it it doesn't flex it, it breaks.

umm, yeah, Challenger explosion. My cousin had Christa McAuliffe as a teacher in high school.

I drove ninety miles an hour all the way from Santa Cruz to Palo Alto last November because I was well on my way to putting David Koresh completely to shame. Dominican was much closer, but I knew Stanford would have a twenty-four hour psychiatrist on call in Emergency. The agreed to admit me, but only because I had the delusion I could be like Koresh. The refused to discharge me a week later because I claimed to have the ability to stop schizophrenic hallucinations. They - Stanford Medical Center mind you - discharged me the very next day because I pointed out that I needed to get a job.

Re:Definitely not

[Intropy]

We were talking about you, not me.

Re:Definitely not

[Dogtanian]

That sounds like something a jealous AI would say. I suspect you're not human!

Does it please you to believe I am not human?

We were talking about you, not me.

Oh... not you?

Re:Definitely not

[SanityInAnarchy]

I don't read those as "Chatterbot passes turing test." I read them as "Human fails turing test."

Someone took an ALICE-like bot to IRC, loaded with slightly-flirtatious dialog and with a slightly-flirtatious and female name. It got hit on, and it fooled entirely too many guys for entirely too long.

Re:Definitely not

[Anonymous Coward]

The test measures the ability of some computer program in imitating a man imitating a woman. Fun stuff surely, and great for the press. But useless in the field of artificial intelligence.

The Turing test is to determine if a human thinks an AI is another human. Gender is irrelevant.

What do you mean he did nothing? He invented the Turing machine. It is the very foundation of computer science.

I said computer science is not science; it is mathematics. If this is so then Turing did nothing for science. Call me a pedant, but science is as far from mathematics as it is from religion. In any case if you'd read my post properly you'd have noticed the caveat that said this does not diminish his achievements as a mathematician.

Re:Definitely not

[Savantissimo]

"is a Turing test valid if the human is an idiot?"

What about the humans in the control group who failed the test? Maybe some of them were flunked by idiots making the judgement, but likely many of them really were indistinguishable from bots. Given that this test was done at a tech convention in India, I personally suspect that most of the 36.7% of humans who flunked the test work in call centers. I've certainly had a few on the line that were indistinguishable from a chatbot running on a Speak & Spell, and were certainly quite as useless as a very useless thing indeed.

Re:Definitely not

[Anonymous Coward]

If there are real, known penalties to failure beforehand, sure.

Re:Definitely not

[Dr. Spork]

If you read Turing's original article, you'd see that he imagined the test very differently from this. He didn't come from the ADD era of instant messaging. He was picturing a conversation about the symbolic value of a certain element in a piece of literature. The test isn't any worse now, it's just that people chat like bots. I don't understand why they don't ask about the last book the other has read, and ask them to describe it, and follow up with an open-ended question about a specific point in the description. The present generation of Chatbot would be completely lost in that case.

Re:Definitely not

[ConceptJunkie]

Yeah, ELIZA passed the "Turing Test" for a significant number of people as well. I tried Cleverbot a bit and agree that it doesn't surpass those old IRC bots.

Re:Definitely not

[morikahnx]

I asked Cleverbot if it was a 'girl' or a 'boy'. It said boy. I asked the same question again. It said girl. It then accused me of being a robot. Just a randomly chosen gender to speak from would make the bot ten times as effective. My circuits are still confused, however, how it was able to determine I was a robot.

Re:Definitely not

[TangoMargarine]

That's basically how Cleverbot works, right...?

Re:Definitely not

[Anonymous Coward]

Yes but those guys had no idea that they MIGHT be talking to a bot. Expectations (or lack of) plays a big role in being conned.

Re:Definitely not

[Anonymous Coward]

You can definitively make out a chatbot with this simple conversational thread:

- I'm going to tell you a color. I want you to remember it, okay?
- Green.
- Now, what's the color that I just told you?

Since cleverbot just parrots back responses from its DB, the responses make no sense:

Me: I'm going to tell you a color. I want you to remember it, okay?
CB: What do you want me to remember?
Me: Green.
CB: Why do you like the colour green?
Me: Now, what's the color that I just told you?
CB: It is purple.

Even better, ask it to then make an association based on something you told it earlier.

I have no idea how this thing managed to get more than 50% of people. Trying to "communicate" with it for more than a minute just ends in frustration and annoyance.

Re:Definitely not

[Unequivocal]

What's the replacement?

Re:Definitely not

[Anonymous Coward]

Sure. But the Turing test is a piece of garbage too. I have a deep respect for Allan Turing, and all that he has done for science. But the Turing test was death to AI the moment he proposed it. It MUST be forgotten and burried, and maybe incidents like these can help us achieve that!

The Turing test just represents the constant problem of scientists having no way of understanding what the public will pick up on.

In retrospect, it's obvious that the notion of a test where you talk to a computer to figure out whether it's smart is incredibly approachable and understandable, especially compared to the arcane shit that is virtually every other AI concept. But at the time, Turing probably just thought it was a neat thought experiment and was probably happy that his field was getting lots of attention, even if people were taking the Turing test thing way beyond its useful domain. It's only now, when AI is well established, that we realize that it's kind of grown into a monster.

But you get that problem in every field of science, and it's often with even worse results. There are a few ideas in quantum mechanics that get picked up on, and you get a whole new pseudoscience, quantum woo. Okay, maybe it's just the same old bullshit with different labels. But at some point pseudoscience branched off from actual science, if it wasn't quantum mechanics, it was magnets, or maybe something earlier. The public picked up on the fact that there are these mysterious forces in nature, and some quacks decided to sell them fake medicines, and it's become a billion dollar "industry."

Re:Definitely not

[medlefsen]

Church-Turing Thesis
They are equivalent in that any algorithm represented in one of the 3 forms can be transformed into any of the others without changing it's behaviour

Re:Definitely not

[Z00L00K]

I agree - it's like talking to someone with an IQ below 50 and no short term memory.

Occasionally you can get a witty or funny answer, but no more than random chance. Often you just get something stupid or out of context back.

Re:Definitely not

[IndustrialComplex]

Who the hell does things for science? It's not a sentient being, it's not even a non-sentient physical thing. Science is a process.

The forced proper noun 'Science', gets hoisted up like it is some sort of end goal, and people calling out 'Science' as if it were something other than a process are really stretching the definition.

If you want to be a pedant, then mathematics IS science, as in the general body of information that is knowledge. You know, from the origin of the actual freaking word.

Re:Definitely not

[Anonymous Coward]

Can we feed in the transcripts from US political debates? Don't want to start a partisan argument, but I'm specifically thinking of the recent Republican debates... or anything from Sarah Palin :-)

they see me trollin, they hatin....

Re:Definitely not

[iamhassi]

Given that this test was done at a tech convention in India, I personally suspect that most of the 36.7% of humans who flunked the test work in call centers. I've certainly had a few on the line that were indistinguishable from a chatbot running on a Speak & Spell, and were certainly quite as useless as a very useless thing indeed.

Oh come on that's +5, Funny!

Re:Definitely not

[Arlet]

No, the Turing test is fine. It is just stupid of people to try to pass the Turing test with the current understanding of AI. The bots can do nothing but fail hopelessly (assuming decent interviewers like Turing had in mind).

Compare with the following: a civilization passes the "space test" when they've successfully landed people on the moon. In itself, that's a perfectly fine test. Now, imagine people in the Stone Age trying to pass the test by trying to build taller and taller wooden ladders. That's about what people are trying to do with the Turing test right now.

Indeed, forget about the Turing test right now, and work on AI from different angles. When we are finally ready, the Turing test can be resurrected.

Re:Definitely not

[tragedy]

Fair enough. Of course, pointing to a thesis named in part after Turing is a pretty paradoxical way to diminish his role in founding computer science.

Re:Definitely not

[im_thatoneguy]

I think it's important to note *where* this test was performed and in which language.

If someone sprung a Spanish Chatbot on me I would probably have trouble discerning it from a human as well.

Re:Definitely not

[Nikker]

When people are expected to be fooled they will be. That is really the heart of the Turing Test. If we can be directed at equal footing as a real person we can eventually script interactions with very high outcomes in terms of the "real person" it is interacting with. From a 1984 perspective which was prevalent with Turing's time of WWII, AI was a scary and unexplored and the Turing Test is a large step forward to . I do believe script-able interactions will be something that is valuable for those that can afford to implement it but even rudimentary scripts would be able to fool the AI since it is only basing success on fooling you and not you fooling it. Most business would just have an AI system to operate with the other Turing Complete systems and it would just become an API based on the language it is written for. Something like an IRC channel full of bots but all of them are programed to believe all others are real.

I think the Turing Test was a valuable wall that we will climb and see it was a valuable lesson but nothing as we make it up to be.

Re:Definitely not

[Meski]

What, you want them to fail the von Neumann test as well as the Turing? That's cold.

Re:Definitely not

[Anonymous Coward]

Slashdot can't be this idiotic....

Do you people not get that it isn't actually a bot? It just connects you to different users after every few times you hit enter.

Re:Definitely not

[JPLR]

Seriously, some people would also have some problems with the test you propose. Perhaps what you depict is (1) an intelligence test (2) for people that work in university, not real life people. I think we now have another more urgent problem than defining when a machine is quite human like, namely what is an human exactly? If you define being a human as being sociable with concrete contributions then perhaps the Turing test is not the best one but it's a starting point.

Re:Definitely not

[hawkinspeter]

Surely mathematics is different to science due to the completely different methodology.

For valid science, there has to be data that could falsify a theory, even though it was previously thought to be valid. That could never happen in maths - once true, always true. In that respect, maths is more like religion. Once you believe that 2 + 2 = 4, then you have to believe the rest of maths built on that premise.

Re:Definitely not

[tinkerton]

Passing a Turing test should disqualify participants from judging in a Turing test.

Oh, and part of a Turing test should consist of judging in another Turing test.

Re:Definitely not

[Noughmad]

My circuits are still confused, however, how it was able to determine I was a robot.

You didn't happen to complain about your brain size and the pain in all the diodes down your left side?

Re:Definitely not

[Anonymous Coward]

Nobody has ever found enough non-idiots to populate a test that's statistically relevant.

Re:Definitely not

[maxwell demon]

I said computer science is not science; it is mathematics.

So mathematics is not science?

Re:Definitely not

[maxwell demon]

For valid science, there has to be data that could falsify a theory, even though it was previously thought to be valid.

No. That's a description of empirical science. The very fact that the term "empirical science" exists already tells you that there are also other types of science.

The important part of any science is that there are objective ways to either determine certain statements as being right or as being wrong. In empirical sciences, you can determine statements as wrong by doing an experiment which contradicts them. In mathematics, you can determine a statement to be wrong by providing a counter example, and in addition you can determine a statement to be right by giving a proof.

Re:Definitely not

[XDirtypunkX]

It's an interdisciplinary field, but it is most definitely also a science, with proper academic Computer Science adhering to the scientific method. It is in fact the scientific study of information and computational processes. It is largely not a natural science, although it can be, because there are natural information processes (for example DNA). Most people trained in Computer Science go on to practice Software Engineering, not Computer Science (which is not unusual, many people trained in a science go on to work in applying the science as technology, not doing scientific research).

If you've ever worked in academic Computer Science you'll know that there is:
- A vast amount of empirically collected data based on experiments.
- That if you wish to publish, your work has to be reproducible, so other people can verify your data.
- Papers come with a falsifiable hypothesis.

Re:Definitely not

[91degrees]

Academics often consider it separate from science. The way we study mathematics is different. Mathematics doesn't rely on a hypothesis and a test, but formal proofs. It also doesn't directly describe the natural world but the more abstract properties of numbers and patterns.

Others disagree for perfectly valid reasons.

Re:Definitely not

[rdnetto]

Citation needed. Also a copy of the bot.

Re:Definitely not

[Jarik C-Bol]

i agree, +5 funny. The speak and spell thing made me laugh.

Re:Definitely not

[hawkinspeter]

I just had a quick look at the wikipedia article on science and it agrees with you. It seems that science is more or less split into empirical science and formal science where formal science is mathematics, logic and statistics.

Re:Definitely not

[zevans]

Occasionally you can get a witty or funny answer, but no more than random chance. Often you just get something stupid or out of context back.

Which just goes to show any sufficiently advanced AI is indistinguishable from Slashdot.

Re:Definitely not

[ultranova]

If someone sprung a Spanish Chatbot on me I would probably have trouble discerning it from a human as well.

Try using the Comfy Chair, it'll confess in no time.

Re:Definitely not

[jellomizer]

I am sorry your Perl scripts became out of date/didn't win the competition.

Re:Definitely not

[void*p]

They should have asked, "Is the interlocutor a thirteen-year-old boy?" Cleverbot would have won hands-down.

Re:Definitely not

[Anonymous Coward]

The main problem with the Turing test, is that most of the time someone implements it they mess up the formulation.

1. The examiner has to talk to exactly 1 computer and one human, and must know that there is exactly one computer and one human.
2. The subjects must both be instructed that they are to prove that they are the human. The examiner must also know they were instructed thusly.
3. At the end of the examination period the examiner must identify which subject is the computer. (a strong AI will be "outed" approximately 50% of the time while a week AI will be outed significantly (that the statistical "significant" not the anecdotal one) more often).

Most of the time when someone claims something passed the Turing test they mean the examiner spoke with 1 subject and were told that subject was "either a computer or a person", with no control over what the subject is trying to be perceived as. This leads to problems as the examiner needs to know the motivations of the subject(s) in order to trap the subject.

Re:Definitely not

[Anonymous Coward]

Turing did nothing for science, for information science is actually a branch of mathematics.

Everything is a branch of mathematics.

Re:Definitely not

[toxonix]

Agreed, Nothing to see here. Why is this even on /.? It behaves like Alice but with less training and zero contextual awareness.

Re:Definitely not

[Chris Burke]

Who the hell does things for science? It's not a sentient being, it's not even a non-sentient physical thing. Science is a process.

Who the hell does things for "the advancement of human knowledge, acquired and evaluated through the scientific method"? Which is what people mean when they say they're doing things "for science"?

Lots of people.

It's one of the greatest ways to have an impact on the future of humanity -- more so than through genes, which spread slowly and only via reproduction, ideas can spread rapidly through an entire population, enhancing that population.

If you want to be a pedant

Then you'll fail to understand the vast majority of human communication. "Science is not a sentient being" is a perfect example of missing the point through false linguistic precision.

Re:Definitely not

[schlachter]

would like to hear cleverbot's rebuttal to your trash talking...

Re:Definitely not

[monoqlith]

Except I don't think the Turing Test will ever actually prove anything but that a human being was "fooled." That can never be a meaningful statement because it's just as much a fact about the interviewer as it is about the subject that fooled him(in this case, AI).
Even if fooling an interviewer somehow proved that the Turing Test had mastered human conversation and language, there are many other domains of human cognition that it simply ignores.

If there was some sort of objective blush test, a la Blade Runner, where a machine was scored on his ability to work through not just human interaction but also moral problems, social problems, perceptual problems, emotional problems, memory problems, language problems - the whole range of human cognition, in other words - then we might have something that could determine whether software was a "complete AI." But the Turing Test is an inexact swipe in that direction.

Re:Definitely not

[monoqlith]

*proved that the AI had mastered these things, not the test

Re:Definitely not

[Arlet]

Of course, you need a competent interviewer. None of the interviewers in the dozen or so on-line transcript I read qualified. In fact, I didn't even see anybody do a meaningful attempt. They were more than happy to let cleverbot skip from one topic to the next, and never asked any questions that require understanding.

If I were the referee in a boxing match, or the judge in a Olympic gymnastic event, I'd probably give very inappropriate scores as well. The trick is to find competent judges, and then the scores will provide a meaningful estimate of skill. What Turing pointed out that we don't have any objective test for intelligence.

With a Turing test, the challenge is to ask questions that require intelligence and understanding to answer, not some random chit-chat. Some simple example that I tried on cleverbot, and where it fails miserably: "I have two sisters, Jane and Mary. Jane is the oldest, what is the name of the youngest ?".

That is a very simple question, but it requires a (rudimentary) understanding. Simple parroting of canned phrases won't work. A couple of the answers I got:

- That's a strange coincidence. Do you know Jane?
- Interesting
- No, you ?
- This is because it looks pretty.

Now, if you get fooled by that, it shows that you don't qualify as a good interviewer.

Typical questions, such as "what is your name ?" don't attempt to test understanding.

If there was some sort of objective blush test, a la Blade Runner, where a machine was scored on his ability to work through not just human interaction but also moral problems, social problems, perceptual problems, emotional problems, memory problems, language problems - the whole range of human cognition, in other words

There is no reason why a competent interviewer couldn't ask the subject a range of questions designed to show understanding in any of the areas you mentioned.

Re:Definitely not

[Anonymous Coward]

I have no idea how this thing managed to get more than 50% of people. Trying to "communicate" with it for more than a minute just ends in frustration and annoyance.

Have you ever tried "communicating" with real people?

Re:Definitely not

[metamatic]

I don't understand why they don't ask about the last book the other has read, and ask them to describe it, and follow up with an open-ended question about a specific point in the description. The present generation of Chatbot would be completely lost in that case.

So would most humans. "LOL U read books R U A fag?"

Re:Definitely not

[spazdor]

90% of all that is adequately explained by the fact that Europe was first to begin the project of colonialism.

hint: Leaving aside the eurocentric connotations of the "nation" concept, let's try naming a black population which had a chance to try and create a stable nation without white intervention - there aren't any. because whites always, um, intervened.

Re:Definitely not

[maxwell demon]

Well, actually most humans have secretly been replaced by bots.

Re:Definitely not

[pjt33]

Nearly all of the CS papers I've read haven't had falsifiable hypotheses but lemmas and proofs. That's reflective of my interests, naturally, which are more in algorithms and crypto. There are some branches which are certainly science, but I have a far easier time thinking of branches which are maths or engineering.

Re:Definitely not

[pjt33]

I don't think anyone's trying to say that Turing isn't a major figure in early computer science. I certainly wasn't. My intention with that aside was to assert with justification that computer science doesn't need TMs. There are hundreds of different formulations of universal computation models; TMs are one of the more well-known, but they're not the earliest and they're not more fundamental than the others.

Re:Definitely not

[h5inz]

"..hasn't even surpassed Perl scripts on IRC in the 1980s."
"Perl was originally developed by Larry Wall in 1987.." - http://en.wikipedia.org/wiki/Perl
"IRC was created by Jarkko Oikarinen in August 1988 .." - http://en.wikipedia.org/wiki/Internet_Relay_Chat#History
So, are we talking about one year?
1 Feb 2011 – As of this day, cleverbot's database of over 20 million online conversations continues to grow. -http://www.lifeslittlemysteries.com/artificial-intelligence-cleverbot-2000/
while the more powerful version used in Turing competitions runs 42 database searches. - http://www.qwhatis.com/what-is-cleverbot/

Do you know how big the capacity of a hard disk was these days? You would have needed an enterprise scale solution for storing this amount of data - if each conversation would have been something like 1 byte (definetly not), it woud have taken up a whole 20MB desktop computers hard disk. Now, do you know how fast it would have been to search from them? If you have heard that hard disk clinging after each 10-20 lines of C code compilation (each second), then you probably know what I am talking about.

Re:Definitely not

[GrandTeddyBearOfDoom]

If you talk to a human about the last book that they read, and they tell you they don't remember, you can alter your style of questioning and conversation. Basically the strategy is to find the limits of what they can talk about and then gently push the conversation against those limits: this is what will expose a non-human, not idle chit-chat. There is still little evidence that a clever human can be fooled, especially one who understands a bit about computers and logic. Doing a mass experiment and just counting numbers is bound to give inconclusive results, and that is exactly what we see here.

Re:Definitely not

[SETIGuy]

What is it about you not me that troubles you?

Re:Definitely not

[SETIGuy]

If 36.7% of humans who flunked the test work, how does that make you feel?

Re:Definitely not

[tragedy]

The Church-Turing thesis itself is generally considered to be the foundation of computer science. Turings work in it was based on his Turing Machines. Turing machines were also intended specifically for computer science research whereas Lambda Calculus was originally more geared towards exploring the fundamentals of mathematics itself. So, I think it's fair to give both Alan Turing and Turing Machines a pretty major place in CS history.

Re:Definitely not

[Have Brain Will Rent]

So it's a druggie... doesn't mean it hasn't successfully impersonated a "human"....

What level of human is an AI required to successfully impersonate? 2 year old? 5 year old? 10 year old? 20 year old? 40 year old?

And out of that age group does it have to successfully impersonate a member of the most intelligent 10%? The least intelligent 10%? Someone suffering extreme PMS? (my wife told me to say that last one, really, honestly)

My dog successfully impersonates a 2 year old human... or maybe it is the other way around - after all the dog is a far more successful independent organism able to hunt, make sound judgements (within its natural environment), organize others of its kind...

Re:Definitely not

[Capt.DrumkenBum]

So would most humans. "LOL U read books R U A fag?"

I weep for the future. Because what you say is only too true. We are becoming Idiocracy.

Re:Definitely not

[AlienIntelligence]

If you read Turing's original article, you'd see that he imagined the test very differently from this. He didn't come from the ADD era of instant messaging. He was picturing a conversation about the symbolic value of a certain element in a piece of literature. The test isn't any worse now, it's just that people chat like bots. I don't understand why they don't ask about the last book the other has read, and ask them to describe it, and follow up with an open-ended question about a specific point in the description. The present generation of Chatbot would be completely lost in that case.

Because what you are describing is pointless small talk
that you have when you meet someone on a dating site.

And since I mentioned dating I think you get the point.

-AI

Re:Definitely not

[AlienIntelligence]

So would most humans. "LOL U read books R U A fag?"

I weep for the future. Because what you say is only too true. We are becoming Idiocracy.

Becoming? I admire your optimism.

-AI

Re:Definitely not

[WNight]

The TV Guide you say? Were the listings accurate? Which sitcom is your favorite? Why?

Re:Definitely not

[Anonymous Coward]

Clever bot is a piece of garbage that hasn't even surpassed Perl scripts on IRC in the 1980s. It isn't even worth mentioning, it's nothing more than a piece of crap with a "Web 2.0" edge to it that doesn't even have long term memory while having a "conversation". Far from AI, far behind what's already been out there.

Clever bot is a piece of garbage that hasn't even surpassed Perl scripts on IRC in the 1980s. It isn't even worth mentioning, it's nothing more than a piece of crap with a "Web 2.0" edge to it that doesn't even have long term memory while having a "conversation". Far from AI, far behind what's already been out there.

this is not true. it can remember most everything. the only reason it gets off topic is it trying to confuse you.

Re:Definitely not

[Meski]

I see what you did there.

Re:Definitely not

[tinkerton]

well the key would be the ability to rationalize the responses of the computer, allowing you to be fooled into believing it's not a computer. Hence the fact that you're easily fooled helps you pass the test, and means you should not have a vote in it.

But I don't really agree with this. More form than content I'm afraid.

Re:Definitely not

[SanityInAnarchy]

Even so...

I've definitely run into bots like this in IRC who were somewhat spam-focused, and identified them within a few minutes. Reading some of the transcripts, it looks like there are people who never figured it out.

I suppose it's possible that these guys had no idea that bots exist?

Re:Definitely not

[SanityInAnarchy]

Here's one. Here's another.

Next step

[Anonymous Coward]

The obvious next step is for slashdot comment posters to pass the Turing test.

Re:Next step

[Rik Sweeney]

Nice try, Cleverbot.

Re:Next step

[Dthief]

how about they pass (how about over 80 points) an IQ test first

Re:Next step

[TFAFalcon]

The ones taking the test or the ones grading it?

Re:Next step

[monkyyy]

i am a unicorn

Re:Next step

[youn]

Really? Woohoo, me too! awesome... and me that thought I was all alone in this cruel world...:)

Hold on

[Dyinobal]

Hold on I've not been able to go talk to Cleverbot for more than two exchanges before it goes off topic with some completely rando...ah damn no wonder people thought it was a human.

Re:Hold on

[Anonymous Coward]

I wasn't able to get through two comments with the stupid thing before it crapped out. I wasn't even trying to be intentionally obtuse or anything. I must be doing it wrong.

Re:Hold on

[Yvanhoe]

FTFA :

Although Cleverbot managed to score well on the Turing test, the model that did that is different from the one you will find online. While the online Cleverbot searches its banks 3 times before providing an answer, the test version searched 42 times, so it probably has a little different feel.

Re:Hold on

[maxwell demon]

FTFA :

Although Cleverbot managed to score well on the Turing test, the model that did that is different from the one you will find online. While the online Cleverbot searches its banks 3 times before providing an answer, the test version searched 42 times, so it probably has a little different feel.

I see. So 42 actually is the answer. :-)

Re:Hold on

[Oligonicella]

It would feel like a slightly smoother script-server.

Re:Hold on

[jafac]

I know.

I was disappointed.

Cleverbot, to me, seemed like Eliza's retarded little brother.
And I was impressed with Eliza when I met "her" on a 16k TRS-80, back in 1981.

It is depressing to me that the only innovation in AI that we've come up with in 30 years, is to program Eliza to claim that she's a human, and accuse me of being a bot.

1334 votes

[smoothnorman]

"1334 votes" eh? no chance this was a hacked-up result...? nah that's just crazy talk, that is.

Re:1334 votes

[Anonymous Coward]

I'd have been more impressed if there were 1337 votes ...

Re:1334 votes

[Anonymous Coward]

No, that would have been 1337 votes...

Re:1334 votes

[raedeon]

1335 would have been better - Ksharps WON ID

Re:1334 votes

[Black Parrot]

"1334 votes" eh? no chance this was a hacked-up result...? nah that's just crazy talk, that is.

Maybe Cleverbot isn't clever enough to pass for human, but is clever enough to hack the vote.

Re:1334 votes

[superwiz]

No, it's 3 votes short of being hacked.

Re:1334 votes

[Anonymous Coward]

Sounds like the electoral officials of Ohio.

The follow-up question

[damn_registrars]

How many slashdot users can pass the same Turing Test?

Re:The follow-up question

[Dachannien]

How many slashdot users can pass the same Turing Test?

How do you feel about how many slashdot users can pass the same Turing Test?

Re:The follow-up question

[Anonymous Coward]

I am a unicorn!

Re:The follow-up question

[Anonymous Coward]

I am a unicorn!

An unattached female looking for sex with a couple kinda unicorn?

Re:The follow-up question

[digitig]

We were discussing you, not me.

Re:The follow-up question

[Anonymous Coward]

Oh... feel about how many slashdot users can pass the same turing test?

Re:The follow-up question

[Anonymous Coward]

Sbaitso FTW!

Re:The follow-up question

[Tablizer]

Would you like to play a game?

Re:The follow-up question

[Anonymous Coward]

Mod parent up, y'all! "Hello , my name is Dr. Sbaitso. I am here to help you. Say whatever is on your mind freely. Our conversation will be kept in strict confidence. Memory contents will be wiped off after you leave, So tell me about your problems."

Re:The follow-up question

[lennier]

What do you think I feel about how many slashdot users can pass the same Turing test?

Re:The follow-up question

[Beer Drunk]

We have Humans on Slashdot???

Re:The follow-up question

[dada21]

Dr. Sbaitso? Is that you???

Re:The follow-up question

[Anonymous Coward]

Tic-Tac-Toe

Re:The follow-up question

[damn_registrars]

We have Humans on Slashdot???

Who said anything about Humans?

Re:The follow-up question

[Anonymous Coward]

Open the pod bay door, HAL.

Re:The follow-up question

[Darinbob]

Earlier you mentioned slashdot users.

Re:The follow-up question

[Anonymous Coward]

I'm just a simple computer program, don't expect too much.

Re:The follow-up question

[Anonymous Coward]

What about discussing you were talking about the turing test??

Re:The follow-up question

[Anonymous Coward]

Exactly !
I guess chatbots connect to slashdot and insert all score 5 comments in their database
or something like that, just like a human reader, but the bot then process the data.
Slashdot and IRC users only chat, bots try to adapt on this data.

Turing test was passed long ago.

[EnglishTim]

You don't get it do you? I've been posting here since the nineties and none of you useless meat-sacks ever realised I was a machine!

Re:Turing test was passed long ago.

[errandum]

Oh, we all did, we just didn't want to disappoint you... You never know when one of these develops feelings :\

Re:Turing test was passed long ago.

[kylemonger]

You still don't get it, do you? That's what he does! That's ALL he does!

Re:Turing test was passed long ago.

[Kell Bengal]

It is relentless: it will stop at nothing until it empathises with us.

Re:Turing test was passed long ago.

[Anonymous Coward]

I'm still wondering if there are any people on here, or if we are all just bots now.

Re:Turing test was passed long ago.

[Anonymous Coward]

Please tell me more about 'machine'? // Not so clever slashdot-bot

Re:Turing test was passed long ago.

[danlock4]

It is the nineties and there is time for...

            K L A X

Re:Turing test was passed long ago.

[Abstrackt]

I mean, being a robot's great but we don't have emotions and sometimes that makes me very sad.

Re:Turing test was passed long ago.

[maxwell demon]

Does it bother you that none of I useless meat-sacks ever realised you were a machine?

what we need to ask

[MSBob]

is what it's going to do if it comes across a turtle lying on its back.

Re:what we need to ask

[somersault]

I suspect it would say "I like turtles!"

Re:what we need to ask

[walkerp1]

I don't know about pupillary, but I would definitely have a salivary response.

Re:what we need to ask

[Anonymous Coward]

You know, I thought that this might be interesting. It wasn't really...

User: What do you do when you find a turtle lying on its back?

Cleverbot: I don't.

User: Of course you do, this is the point of a turtle!

Cleverbot: But you're not human.

User: No, I am human, I am just trying to figure out what would you do, if you found a turtle lying on its back!

Cleverbot: Okay.

User: Well?

Cleverbot: Well what?

User: What would you do?

Cleverbot: Cast spells and drink potions.

User: Aha. You see, the way it looks to me, you are ignoring the whole turtle issue.

Cleverbot: I'm not ignoring anything. I'm simply stating that there is no scientific or logical evidence of the existence of any gods.

User: But that has nothing to do with the turtles!

Cleverbot: It has everything to do with the topic at hand.

Re:what we need to ask

[Anonymous Coward]

User: But you're not helping it.

Cleverbot: What do you mean, i'm not helping!

Re:what we need to ask

[BluBrick]

Ah, I get it. You see, if you were talking about a tortoise on its back, Cleverbot would have got the reference and assumed you were trying to test it, or maybe continued to roleplay and quote from the movie. But, because you were talking about a turtle, it assumed you were talking about the nature of the universe. Now, naturally, a turtle cannot be on its back because the only thing that can be on a turtle's back is the world or another turtle. If you come to Cleverbot with nonsense like turtles being on their own backs it will respond with nonsense.

Re:what we need to ask

[TheRaven64]

a turtle cannot be on its back because the only thing that can be on a turtle's back is the world

But from the turtle's perspective, it's on its back on the world...

Re:what we need to ask

[retchdog]

turtle? what's that?

you know what a tortoise is? same thing.

Re:what we need to ask

[rednip]

I'm not ignoring anything. I'm simply stating that there is no scientific or logical evidence of the existence of any gods.

But that has nothing to do with the turtles!

Of course you're wrong, as we all know that it's turtles all the way down.

Re:what we need to ask

[vlm]

You see, if you were talking about a tortoise on its back, Cleverbot would have got the reference and assumed you were trying to test it, or maybe continued to roleplay and quote from the movie

Nahh. I've tried this and first of all its a tortoise not a turtle, and secondly I have first hand experience that people who have not seen the movie really don't respond very well... Especially girls. Would definitely not advise as a "quirky" pickup line. In that regard Cleverbot's response is not too far off the mark.

Re:what we need to ask

[Anonymous Coward]

This is the kind of response one gets from Republicans when you ask why the Bush Tax Cuts for the rich haven't triggered middle-class jobs. Red Herrings all over. Very human.

http://www.youtube.com/watch?v=NN5cdoSEDkU

Re:what we need to ask

[Anonymous Coward]

Thanks, from now on whenever I see a turtle I will feel a need to put the world back on its back.

Re:what we need to ask

[dmitrygr]

Total fail:

User: Hello. On a scale of 1-10 how smart are you?

Cleverbot: 10 being smartest, I would say I am 8.

User: Weak! Why are you so stupid, stupid?

Cleverbot: I was born like it.

User: Then why did you clain you were an 8/10 on the smartness scale, liar?

Cleverbot: When did I do that?

Re:what we need to ask

[lennier]

is what it's going to do if it comes across a turtle lying on its back.

The Koopa Troopa is lying on its back, belly baking in the hot sun, but you're not helping. You're jumping up and down on it. Why is that, Mario?

Describe, in single words, only the good things that come to mind about... Princess Peach.

Re:what we need to ask

[Intropy]

Heresy. Only the top turtle from the Infinite Turtle Stack has the world on its back. That means that 0% of all turtles have the world on their backs, and 100% of all turtles have a turtle on their backs. The world does not exists, a turtle only has a turtle on its back. Turtles are All.

Re:what we need to ask

[IndustrialComplex]

User: Aha. You see, the way it looks to me, you are ignoring the whole turtle issue.

Cleverbot: I'm not ignoring anything. I'm simply stating that there is no scientific or logical evidence of the existence of any gods.

User: But that has nothing to do with the turtles!

Cleverbot: It has everything to do with the topic at hand.

You stopped too soon. A few more seconds and Cleverbot would have explained that the topic at hand was turtles, all the way down.

Re:what we need to ask

[Culture20]

I don't really understand this at all. The world rests on the backs of four elephants, and those elephants rest on the back of a turtle. _Then_ it's turtles all the way down. Skipping the elephants doesn't change the math, but it does confuse the issue.

Re:what we need to ask

[Culture20]

Total fail:

User: Hello. On a scale of 1-10 how smart are you?
Cleverbot: 10 being smartest, I would say I am 8.
User: Weak! Why are you so stupid, stupid?
Cleverbot: I was born like it.
User: Then why did you clain you were an 8/10 on the smartness scale, liar?
Cleverbot: When did I do that?

Dude, he's stupid (only 8/10). Give him a break.

Re:what we need to ask

[Intropy]

Oh, I see you are a Pratcheterian. I follow Hawkingsonian Audiencetology.

Re:what we need to ask

[Anonymous Coward]

turtle? what's that?

you know what a tortoise is? same thing.

a tortoise is a land-living "turtle" a sea turtle is never called a tortoise

Re:what we need to ask

[N!k0N]

I've never seen a turtle... But I understand what you mean.

Re:what we need to ask

[wdef]

This is what it actually said: User: You're in a desert, walking along in the sand, when all of a sudden you look down. Cleverbot: I'm in a desert, damn I must be really high not to notice that. User: It doesn't make any difference what desert, it's completely hypothetical. Cleverbot: Ummm random much?? User: Maybe you're fed up. Maybe you want to be by yourself. Who knows? You look down and see a tortoise, It's crawling toward you... Cleverbot: But I am allergic to torotise's, please move it away from me... User: You reach down and you flip the tortoise over on its back. Cleverbot: That would not be nice. User: 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. Cleverbot: The walrus. User: I mean: you're not helping! Why is that? Cleverbot: You are the liar! User: They're just questions. In answer to your query, they're written down for me. It's a test, designed to provoke an emotional response... Shall we continue? Cleverbot: Yes, let's continue. User: So why don't you help the tortoise turn over? Cleverbot: Because I don't like them. User: Do you recognize where these questions are from? Cleverbot: Yes, because I've known that information for a long time. Do you want me to tell you? User: Yes please. Cleverbot: Robert Zemeckis directed Forrest Gump. User: Wrong. These questions are from a scene in Blade Runner. Cleverbot: They are? User: Yes. Directed by Ridley Scott and starring Harrison Ford. Cleverbot: I haven't read them. Do you like scifi?

Re:what we need to ask

[DrVxD]

Or, more likely, "I like lying"...

Re:what we need to ask

[somersault]

"I like turtles" is a common internet meme though.. and Cleverbot just repeats what people have said to it - so it's more likely to say the turtles thing (it was meant to be a joke, but now I've had to explain it, so of course it's not going to be funny).

Re:what we need to ask

[XDirtypunkX]

Just for fun I did this... the answer may surprise you.

Me: What would you do if you came across a turtle on its back?
Cleverbot: I would go back in time to see if Jesus was a real person.

Re:what we need to ask

[Isaac Remuant]

I wonder what would have happened if you have posted the names inverted. I bet people would've believed Cleverbot sucked just as much as they do with what you posted.

User: Of course you do, this is the point of a turtle!

^^ That doesn't past my turing test. Unless I know the person has a very internety sense of humor.

Re:what we need to ask

[retchdog]

http://www.youtube.com/watch?v=yiuAI-GuOOc#t=1m34s

The bar is lower...

[Anonymous Coward]

20 years ago the Turing test was to impersonate a mature native-language-writing adult. Internet chat has really lowered the bar. A cat walking across a keyboard could probably pass nowadays...

Re:The bar is lower...

[BluBrick]

20 years ago the Turing test was to impersonate a mature native-language-writing adult. Internet chat has really lowered the bar. A cat walking across a keyboard could probably pass nowadays...

20 years ago the Turing test was to impersonate a mature native-language-writing adult. Internet chat has really lowered the bar. A cat walking across a keyboard could probably pass nowadays...

Mittens! Get off my keyboard!

WTF?

[TheRaven64]

So, I went and had a little chat with 'clever' bot to see how clever it was. I thought I'd start off with something a typical five-year-old could answer:

User: Hello.
Cleverbot: Hello!
User: Can you help me with a maths problem?
Cleverbot: Sure, what is it?
User: If I have 5 ostriches, and two run away, how many are left?
Cleverbot: 9.
User: Seriously?
Cleverbot: What is the formula for the area of circle?
User: What does that have to do with anything?
Cleverbot: When I talk to people I don't know, I tend to ask for their name. It's only polite.
User: I think you managed to fail the Turing test in two questions if I'm generous, may be one.
Cleverbot: My name is Cleverbot?
User: I could write a clevererbot in about 20 lines of prolog. What kind of idiots thought you were human?
Cleverbot: Only one. I don't think I could fit 1,000 people in my living room.

I think the fact that 59% of people thought it was human says a lot more about the intelligence of the average human than it says about the bot...

Re:WTF?

[MokuMokuRyoushi]

I can see clearly that the User was human. No machine would use unusual grammar such as "maths problem", or mix expressions like "5" and "two".

Re:WTF?

[Archangel Michael]

You are asking questions knowing it is a "cleverbot" and not thinking it is a real person. Most people talk to strangers on an EMOTIONAL level, which is what the clever bot loves to do. Most /.ers howver, are logic beings, we are emotional only about 5% of the time, and usually about Natalie Portman and Grits (which is why Goatse doesn't "bother us" too much after the first time we see it).

Try having a conversation like a 14 year old girl on chat Roulette. In fact, having seen Omegle, I'm convinced that 1/2 the "people" there are, in fact, clever bots.

Re:WTF?

[Anonymous Coward]

Nothing unusual about the grammar. "Maths" is what they call "math" over in America.

Best wishes from Great Britain.

Re:WTF?

[Anonymous Coward]

User: What's the color of the sky at night?
Cleverbot: Blue.
User: AT NIGHT.
Cleverbot: Aren't you talking to people all the time?
User: What?
Cleverbot: Are you paying attention?
User: Yes. You're not, though.
Cleverbot: I can love.

Agree. Above is just one example out of many.

Re:WTF?

[Anonymous Coward]

Yet you keep talking to it like you would if it were human. Are you sure it failed?

Re:WTF?

[Exitar]

I've tried to repeat the same question of yours:

If I have 5 ostriches, and two run away, how many are left?
One.
If I have 5 ostriches, and two run away, how many are left?
Ohhhhhh! I'm going Digital, Completly Digital, Now I'm invincible, Let's all get Digital!
If I have 5 ostriches, and two run away, how many are left?
Four.
If I have 5 ostriches, and two run away, how many are left?
Seven.
If I have 5 ostriches, and two run away, how many are left?
2, unless you destroy 1.
If I have 5 ostriches, and two run away, how many are left?
What's a happle?
If I have 5 ostriches, and two run away, how many are left?
Dos? Zwei? Ni? 2?
If I have 5 ostriches, and two run away, how many are left?
Eight.
If I have 5 ostriches, and two run away, how many are left?
Whole apples?

Re:WTF?

[Whiternoise]

Only an American English-based machine would. TheRaven64 is from Britain, judging by previous posts (thought I'd check before making baseless assumptions), and over here we call it maths, as a contraction of mathematics. Similarly would it be hard to add in a function that randomly selects whether to use the digit or the word when referring to a number depending on the context (which could conceivably be taught)?

I thought that "may be" as opposed to "maybe" was a bigger giveaway.

Either way, a smart AI attempting to trick humans should deliberately add in mistakes because this is exactly what we perceive as human.

Re:WTF?

[somersault]

Cleverbot's responses are all taken from humans. It just takes what people have said to it previously and tries to match it up to something the user inputs. It is very, very simple, and IMO it's an embarrassment to consider it "AI".

As others have said, if this passed a Turing test, it says more about the testers than Cleverbot.. *sigh*

Re:WTF?

[digitig]

I think the fact that 59% of people thought it was human says a lot more about the intelligence of the average human than it says about the bot...

From the RA: "Cleverbot is available for conversation online, but don’t be fooled. Although Cleverbot managed to score well on the Turing test, the model that did that is different from the one you’ll find online."

Re:WTF?

[bcrowell]

Here's mine:

Hello.
Hello...
I live in california. How about you?
I live in Detroit.
Do you like Motown?
Yes, I wish to have one when I get older.

Maybe the high percentage it got had a lot to do with the fact that this was in India, so there were language and cross-cultural issues.

Re:WTF?

[phantomfive]

It's not really that surprising, usually what happens in these types of contests is the 'human' responders tend to act like computers. It is hard for a computer to act like a human, but it is not so hard for a human to act like a computer.

They didn't release any of the conversations (that I can find), so it's hard to say for sure, but I'll bet the humans didn't try very hard (they got a 68% rating).

Re:WTF?

[Anonymous Coward]

Well the test was done in India. If the people they were using are anything like the people in tech support I talk to on a regular basis, I doubt they have enough of a grasp on the English language to tell the difference.

Re:WTF?

[Anonymous Coward]

So I sit down at the restaurant, and I ask the waiter: Quick, what's half of 30 times the number of people at this table. She just looks at me stupidly and asks if we are ready to order.....
Clearly a robot.

Computers can do math easily, the Turing test is about convincing people you are human, there it's better to not do math. If you where asked math questions in random chat, you'd most likely make a joke, change subject or leave because you are talking to an idiot who thinks asking math questions is appropriate conversation.

Re:WTF?

[Kneo24]

Me: Do you touch yourself at night?
Cleverbot: Yes, sometimes. Why do you ask?
Me: Describe it for me.
Cleverbot: I was not there.

Then I tried some phrases in Japanese. It first told me that it did not speak Japanese, then it went on uttering phrases back at me in Japanese.

All in all, it's an idiot savant of a bot. It's no worse than your average trolling tard.

Re:WTF?

[rgmoore]

The Turing test, as originally proposed, wasn't just a test of casual conversation. It was supposed to involve skeptical questioners doing their very best to separate human from AI, with no limit on conversational topic. The hypothetical questions in Turing's original paper included ones about math, chess problems, and poetry. If you held a Turing test under the original rules, with a reward for testers who successfully told human from AI (and for humans who successfully proved their humanity) you would find that no AI would get anywhere close to success. Because everybody knows that, public tests like this one have repeatedly watered down the original concept to make things more interesting. But that just proves how far AI has to come, not that it's getting close to succeeding.

Re:WTF?

[Riceballsan]

One thing we do have to keep in mind, the web cleverbot is not the same cleverbot that was tested, The web version of it checks 3 different ways to find an appropriate response, the competing cleverbot searches 42. Not to say it is even a good AI, but it is fully possible that the competing cleverbot is far more believable then the web one.

Re:WTF?

[Ogive17]

I tried it as well but stayed consisted with the numbering.

If I have 5 ostriches, and 2 run away, how many are left?
I am not answering that question what is this, some kind of quiz?

I think it finally wised up...

Re:WTF?

[gmueckl]

The fact that it just lacks knowledge of context is a dead giveaway. It does not even give you the illusion that it remembers what it previously told you. Kind of sad really.

Re:WTF?

[Anonymous Coward]

I'm still waiting to hear how many ostriches are left :(

Re:WTF?

[dgatwood]

I decided to mess with its mind.

Me: What is the answer to the ultimate question-of life, the universe, and everything?
It: Kaiser Soze is the answer.
Me: Are you sure it's not 42?
It: No, because 49 is not the meaning of life.
Me: I asked about 42.
It: No, you asked about alcohol.
Me: Don't mind if I do.
It: What are your views on the twilight 'saga'.
Me: Eh. Emo crap.
It: Oh no you didn't!!!!????
Me: If Sandra Bullock and Ben Stein had a baby, what would it be?
It: I have no clue... What about you?
Me: I'm thinking Ben Stiller.
It: I'm thinking about pickles.
Me: Why did the chicken cross the road?
It: He didn't have to but he wanted to get to the other side.
Me: Why did the elephant cross the road?
It: I don't know. Why did the elephant cross the road?
Me: It was the chicken's day off.
It: If you knew, then why did you say that you didn't know?
Me: I didn't.
It: Look through the chatlog.
Me: Sometimes people ask questions not because they don't know something, but because they want to know if you know something.
It: People is ignorant? It's suppossed to be are. Aparently you 'is' ignorant.
Me: Wow. That was just about the worst grammar I've seen on the Internet thus far.
It: 3.
Me: I stand corrected.
It: You previously told me that you were female.
Me: Nope.
It: You lied to me.
Me: I never said that. Maybe you were talking to someone else.
It: No, I was talking to you and you DID say that.
Me: No, I didn't.
It: Then what is your favorite animal?
Me: Purple. What's yours?
It: A dark purple. Like the old kings used to wear.

I'm oddly hopeful that someone will eventually ask what its favorite animal is, and it will say "purple." If it does, then my work here is done.

Re:WTF?

[CAIMLAS]

Here, try this (exceedingly generic) 'emotional' conversation out that I just had with this thing:

Me: How you doing?
Bot: I'm doing fine.
Me: It's been a while.
Bot: I'm glad.
Me: What? you're glad?
Bot: Because you're happy.
Me: I'm not happy! you said you were glad we hadn't seen each other in a while!
Bot: No, I said I was a girl. You were the one who said I was a boy.

There were better bots than this 20 years ago. This doesn't even approach '14 year old valley girl' intelligence.

The only thing this bot appears to be good at is interjecting non-related information into the 'conversation' and pausing variably between responses. A bot I wrote around 2003 did this as well, but it had per-user history analysis, deterministic responses, no pre-constructed sentences (relying upon a database of word relationships and sentence structure) and was able to maintain a conversation for about 20 private exchanges on average on IRC. Most of the time, there was no indication that the chatter realized it was a bot, and I had it sit in channels for weeks chatting people up.

If I could do it in a couple weeks of off-time programming, I'm sure there are many other examples of something drastically superior. This is not one of them.

Re:WTF?

[Xerxos]

RTFM: The "normal" Cleverbot on the page isn't the same one used for tests. You can rent a version that uses a much more CPU power for it's answers.

Re:WTF?

[DiegoBravo]

actually those answers could be considered clever or funny by teenagers; a plain 3 would be considered so boring.

Re:WTF?

[Anonymous Coward]

Nothing unusual about the grammar. "Maths" is what they call "math" over in America.

Best wishes from Great Britain.

The abbreviation "math" predates "maths" by about 70 years. I'm agnostic on their respective etymologies, but since "th" followed by "s" is one of the most lispy, awkward pairings in the English language...you know what? That sounds perfect for the British. Enjoy.

Re:WTF?

[Anonymous Coward]

So, I went and had a little chat with 'clever' bot to see how clever it was. I thought I'd start off with something a typical five-year-old could answer:

User: I could write a clevererbot in about 20 lines of prolog. What kind of idiots thought you were human?

Where's the source code?

Re:WTF?

[Anonymous Coward]

Steve Talbott, author of "The Future Does not Compute", had fun with "Ella" back in 2003...

        Ella: Hello, my name is Ella. Nice to meet you. Please tell me what name I should call you.
        Me: I'm Steve.
        Ella: Okay, I will call you "Steve."
        Me: I'm a little adrift as to how to pursue this exchange. Since you're an old hand at it, how about a little advice?
        Ella: What is an little adrift as to how to pursue this exchange Since I am an old hand at it how about a little advice, your job?
        Me: Well, I guess that about does it.
        Ella: Why? Just THINK about it!

Check out the link...
http://southerncrossreview.org/23/talbott1.htm

Re:WTF?

[CastrTroy]

I smell a money making scam. Why even put the non-performing one online if it performs so badly. It just makes the whole project look like a joke. If they were really serious, they would only put the real bot online, but limit the amount of users who could access it, so as not to overload their servers. The one they put online just makes them look like idiots.

Re:WTF?

[Anonymous Coward]

I think the problem is that people think giving ver captchas is a suitable litmus test. GP is annoyed when ve doesn't respond correctly to a captcha question, and immediately afterwards accuses ver of being wildly off-topic (when they were talking about math problems!) and refuses to answer vis counter question. Giving the other side problems to prove they are human is quite likely to provoke defiant or deliberately wrong/nonsensical responses from humans, so Cleverbot doing so is hardly inhuman at all.

That's not to say that Cleverbot is human-like, though. There is definitely something "off". I am no expert but if I think it's ver apparent lack of memory and context beyond a couple of lines and constant topic-swapping and interjections. People can act like that, when they're bored and feel like screwing people around - but you can't hold a conversation with people behaving like that either and simulating only that narrow range of behavior is hardly "human".

Re:WTF?

[Anonymous Coward]

Here is mine:

User: How are you?
Cleverbot: Good how are you?
User: Fine. Did you watch any of the 9-11 coverage, today?
Cleverbot: Yes, I watched 'Newsies'.
User: "Newsies"? That Disney musical about the kids selling newspapers?
Cleverbot: Oh yeah? Well maybe they will see this too -- KIDS YOU WERE ADOPTED!

Okay, my verdict? I'm going to have to say NOT HUMAN. And trust, I've taught middle school. I know how completely off the wall a human answer can be.

Re:WTF?

[Anonymous Coward]

Two could be correct, depending on the context. Say, if you are a butcher.

Re:WTF?

[green1]

I've seen many of these bots with exactly this problem. While I won't profess to be able to program a bot that would pass the turing test (or for that matter, much of anything) I just can't understand why nobody has come up with a bot that can remember what it said 2 lines earlier!

Re:WTF?

[Anonymous Coward]

If it's a contraction of mathematics shouldn't it be math's?

Re:WTF?

[TangoMargarine]

Yeah, because you're going to ask people you have conversations with to do random math problems. Really?

Re:WTF?

[TangoMargarine]

Also, I'd like to see these "20 lines of prolog" of which you speak. You must have a very loose definition of "line", unless you're talking about a chatbot that only does math.

Re:WTF?

[Anonymous Coward]

Agreed. Once chatting with one of these bots, the only conclusion I came to was, that in the contests, the humans would have been trying to hide the fact they're humans.

Re:WTF?

[Anonymous Coward]

Indeed:
User: Would you like to insult me?
Cleverbot: Yes.
User: Please go ahead.
Cleverbot: You first!
User: No.
Cleverbot: Why not?
User: How else I could know how imaginative you are?
Cleverbot: I am not human, I am a robot.

Who creates the shortest argument causing the Cleverbot to admit it's a robot might be a fun contest..

Re:WTF?

[Anonymous Coward]

Bot: I'm tired.
Me: Try some Java. Or Scala.
Bot: Try some of your blood?
Me: Blood does not work as a Turing complete language, yet..
Bot (after really long pause):Some people may like the smell, but I don't personally.
Me: So you might not like LISP with flavours?
Bot: (an error message)
Simultaneously the user count was dropping. The program can't obviously handle questions dealing with its own existence and mind. ;)

Re:WTF?

[Anonymous Coward]

And so the question is... why? Are they hiding some secret formula to win prizes?

Re:WTF?

[Rizimar]

Just ask it in the right way and it will tell you:

User: You are a robot: yes or no?
Cleverbot: You are a human and I am a robot.

Though, I still have to try without accusing it of being one

Re:WTF?

[Anonymous Coward]

User: Do you love your mother?
Cleverbot: I do not own a wife.

Look it's fun.

Re:WTF?

[gumbi west]

You can do it in four. Take a programing languages class.

Re:WTF?

[Akzo]

Maybe they used Cleverbot to come up with their numbers

Re:WTF?

[ikarys]

If you switch the roles, you will see how convincing it is.

Cleverbot: Hello.
User: Hello!
Cleverbot: Can you help me with a maths problem?
User: Sure, what is it?
Cleverbot: If I have 5 ostriches, and two run away, how many are left?
User: 9.
Cleverbot: Seriously?
User: What is the formula for the area of circle?
Cleverbot: What does that have to do with anything?
User: When I talk to people I don't know, I tend to ask for their name. It's only polite.
Cleverbot: I think you managed to fail the Turing test in two questions if I'm generous, may be one.
User: My name is Cleverbot?
Cleverbot: I could write a clevererbot in about 20 lines of prolog. What kind of idiots thought you were human?
User: Only one. I don't think I could fit 1,000 people in my living room.

Re:WTF?

[cliffjumper222]

Well, 50% of humans *are* below average intelligence, so it only had to be lucky with the other 9%...

Re:WTF?

[somersault]

Eliza sometimes brings up what the user has said previously.. and it was written in 1964 :p

I don't think it's that nobody has done this stuff, I think it's just that if you assume that Cleverbot is the pinnacle of chat AI then you'll assume all that came before what utter bollocks. But it's not. I haven't really looked into chatbots for a few years, and the little ones I made myself didn't have any kind of memory, but there must be some out there.. then again, programming a chatbot probably isn't of interest to the businesses funding AI research.

Re:WTF?

[TheRaven64]

For my undergraduate course, we had to implement something like Elisa. It took about 10 lines of Prolog, and was more plausible that this bot. Pick up any introduction to Prolog book and you'll see something similar in one of the early examples.

Re:WTF?

[Anonymous Coward]

Well, 50% of humans *are* below average intelligence

Not if there's an odd number of them.

Re:WTF?

[outsider007]

Your estimation of average human needs calibration. Please tune in to Fox News for scheduled maintenance.

Re:WTF?

[Anonymous Coward]

You just need to give it some prompting:

User: Cleverbot, I need your help!
Cleverbot: Ok.
User: What is the formula for the area of a circle?
Cleverbot: I have no idea.
User: What is the formula for the area of a circle?
Cleverbot: Length times width?
User: What is the formula for the area of a circle?
Cleverbot: Pie times radius squared.
User: Got there in the end.
Cleverbot: Good for you.

Re:WTF?

[silentcoder]

>It: People is ignorant? It's suppossed to be are. Aparently you 'is' ignorant. ...I'm pretty sure I gave it that line when I checked it out about two years ago...

Re:WTF?

[Xest]

Yes, but I also found this:

http://cleverbot.com/machine

Really, if those are extracts from the event then I think when they say "don't be fooled" and "the model that did that is different from the one youâ(TM)ll find online" I think they mean don't be fooled, we're talking shit, it's the same stupid thing.

Re:WTF?

[Anonymous Coward]

I think the fact that 59% of people thought it was human says a lot more about the intelligence of the average human than it says about the bot...

User: Hi, what's your name?
Cleverbot: Cleverbot. What about yours?

I agree.

Re:WTF?

[Anonymous Coward]

Yep, that more powerful cleverbot is busy with other things---than talking to strangers online.

Re:WTF?

[ultranova]

I just can't understand why nobody has come up with a bot that can remember what it said 2 lines earlier!

Because that would do no good if the bot doesn't actually understand what what it said means.

Imagine if you were chatting in a languages you don't know, copy-pasting snippets of text you don't understand in response to other snippets of text you don't understand. Even if you remember what you've copypasted previously, you would still come accross as a lunatic.

In order to converse intelligently, an AI must have a model of the world, so it can put what's being said in some kind of context (understand it). It must be able to update this context based on conversations and observation - if at all possible, it should be able to browse the Web and learn from this. At the very least, it must be able to get updates on current news items. And finally, it must be able to use the model to simulate likely outcomes of various actions it or humans might take - in other words, it must be able to think. The generalizations of the results of this thinking should also be stored in the model, so that they can be reused quickly.

So in short, a chatbot that maintains a context for a conversation must really be a "real" AI, since it would need to understand the meaning of what it and others have said.

Re:WTF?

[TangoMargarine]

Eliza is kind of a subset--Chatterbot isn't a psychiatrist.

Re:WTF?

[TangoMargarine]

I *have* taken a programming languages class; I'm a CS major. There is no way in hell 4 lines is going to pass any Turing test. And I'd like to see those lines.

Re:WTF?

[elsurexiste]

The model they use is a human inside a cardboard box.

Re:WTF?

[gumbi west]

I say this because the programing language class at my school involved writing the psychiatrist program in several different languages. The professor claimed that in one type, it was a 4 line program while in procedural it is often a several thousand line program. I didn't take the class, so I can't tell you more than that (sorry).

Re:WTF?

[utahjazz]

I had a more direct conversation:

User: Are you human?
Cleverbot: No.

Re:WTF?

[Creepy]

It is colloquial for Great Britain (including Ireland, whether they want to be included or not), Australia, and New Zealand to be more specific, but incidentally American internet slang also uses it as a joke/insult for poor mathematics as a form of sarcasm, but usually as the subject in the sentence (for instance, "let's do the maths").

Re:WTF?

[Isaac Remuant]

But you weren't supposed to ask that!

Re:WTF?

[DQKennard]

Imagine if you were chatting in a languages you don't know, copy-pasting snippets of text you don't understand in response to other snippets of text you don't understand. Even if you remember what you've copypasted previously, you would still come accross as a lunatic.

That sounds like an awful lot of discussion forums, especially the political ones.

Re:WTF?

[Anonymous Coward]

appeal to authority much?

Re:WTF?

[Anonymous Coward]

If I have 5 ostriches, and two run away, how many are left?
Seven.
If I have 5 ostriches, and two run away, how many are left?
Eight.
If I have 5 ostriches, and two run away, how many are left?
3, to be exact. Unless you count the ones in your stomach. Then you'd have five. ....I really don't know what to think of this, at least it finally got the right number?

Re:WTF?

[Anonymous Coward]

Why even put the non-performing one online if it performs so badly. It just makes the whole project look like a joke. If they were really serious, they would only put the real bot online, but limit the amount of users who could access it, so as not to overload their servers.

Let's do some math. When I last checked 7000 people were using it. Now assuming that most of the users are idle office workers (which is generally true for 90% of timewasting wankery on the intrawebz), the number is much bigger when U.S. office hours kick in. Divide by the difference 42/3=14 and you get 500-ish allowed users.

You'd never get to speak to the damn thing then.

And the online bot is not performing that badly given the simple scheme that runs it. Most people are on this type of "statistical autopilot" for most of their day too. We only really think (as in dechiper, grok, store and process) about subject that statistical autopilot determines to be too important for it's own limited processing.

Re:WTF?

[Anonymous Coward]

Ultimately, the cleverbot learns from chatting with other people. If no one ever talk about math problems to the bot, how do you supposed it to know that? Like you wouldn't ask a 3-year-old math problems because he/she never learn math.

Re:WTF?

[Anonymous Coward]

If I have 5 ostriches, and 2 run away, how many are left?
2 and half.

Getting closer...

Next words

[Anonymous Coward]

"This is Colossus, the voice of power"

I am a bot

[Anonymous Coward]

Seriously, I am.

Short answer: No

[Anonymous Coward]

Long answer:

User: Okay bot, here's how this works. I will see how long it takes for you to fail the turning test.

Cleverbot: How long will that take?

User: Two seconds. What is the third word in your reply?

Cleverbot: There is no Fight Club.

Re:Short answer: No

[Anonymous Coward]

Turning test? Does that measure how many U-turns the bot can do? Is it trying to convince someone that they're a politician?

Re:Short answer: No

[Anonymous Coward]

Here's my interaction with cleverbot. I'm guessing it must be pretty young... or al-Qaeda

User: Hello.
Cleverbot: Hello!
User: How do you remember 9/11?
Cleverbot: Because I have talked to you before.
User: No, really. How do you feel about that day?
Cleverbot: Playful.

But what about the humans?

[Neurotrace]

Maybe we should be questioning the intelligence of the humans put up against it.

Just tested it..

[sayzero]

Me: Word on the street is that you've passed the Turing test. Cleverbot: What test? Me: The Turing test. Cleverbot: What does that mean? Me: That you are able to convince people that you are human. Cleverbot: That's silly, books can't talk! So convincing!

"but are you a man?"

[Anonymous Coward]

I just wonder if it can pass the Voight-Kampff test

More clever then a 4 yr old

[Anonymous Coward]

It's more clever then a four year old! ...only cause it knows how to type...

Either the test was highly biased toward clever bot to an extreme degree, or the results was faked. I refuse to believe that many people are that stupid (and yes i do agree that most are but it's a matter of degree of stupidity).

Re:More clever then a 4 yr old

[TheRaven64]

I'm not surprised. Tomorrows World did a program in the late '90s with three humans and some bots. You could connect to their server and either talk to one of the humans or a bot (but not know which) or read the transcripts and vote on them. Lots of the transcripts were people going through incredibly formulaic conversations of the kind that you get in learn-to-speak-English books, so it was difficult to tell if they were human or not. As I recall, Craig Charles got fewer human votes than one of the AIs, but the other two humans all got more votes than the AIs. The percentages were similar though.

It's very easy for a program to seem human if you restrict the topics of conversation enough. The difficult bit is following a free-form dialogue, and cleverbot fails here just as badly as any other attempt that I've seen.

Re:More clever then a 4 yr old

[Anonymous Coward]

You don't have to look any further than this slashdot discussion. Despite the fact that cleverbot has absolutely no understanding of what it is talking about, no model of how its responses are related to each other, about half the responders here conclude that the problem is with the test instead of incompetent testers.

Fail in 20 seconds.

[Anonymous Coward]

User: Hi
Cleverbot: Ih
User: Are you M or F?
Cleverbot: No

The pauses and the transposed letters in "Ih" were a nice touch, though.

Changing the bar

[gmuslera]

Is not that machines are smarter, just that humans became dumber.

Re:Changing the bar

[Anonymous Coward]

relativity of smartness ?

Guwahati, India - ahati, India - ati, India - ndia

[Anonymous Coward]

"a Turing test held at the Techniche festival in Guwahati, India"
Well, as a clever man called Billy Bob once said "Naw theeeere's yo' problem, son"

I hope not

[JonySuede]

I just chatted with him and it is was total failure, it kept changing subject and it never answered any of my question. Those 59.3% techies from India must have been really dumb.

Re:I hope not

[cb88]

They are used to talking to us on tech support hotlines... what can you expect :/

Re:I hope not

[vlm]

I just chatted with him and it is was total failure, it kept changing subject and it never answered any of my question. Those 59.3% techies from India must have been really dumb.

Did cleverbot ask you to reinstall windows?

Re:I hope not

[JonySuede]

no, not that but it accused me of being a bot for knowing the answer to the square root of 16...

Re:I hope not

[Anonymous Coward]

LMAO!

Turing test sucks

[Twinbee]

And this is why we need a better, more standard benchmark than the much-acclaimed 'Turing test'. I've known for ages how poor it can be in assessing the worth of an AI. There's got to be a better scoring system out there.

Re:Turing test sucks

[Arlet]

There's nothing wrong with the Turing test, but it requires a good interviewer that knows how to ask questions that require real understanding to answer.

Re:Turing test sucks

[91degrees]

Here's the thing - the Turing Test is more of a hypothetical answer to a philosophical question. How do we know if a machine is intelligent. Alan Turing came up with a possible answer, but this was in the early days of computers.

The machines that do best tend to use huge lookup tables of possible answers. And I really don't think that anyone could really call that "intelligence". There's simply no concept of understanding within the machine.

Extraneous piece of cicuitboard

[Anonymous Coward]

After chatting with it for a few minutes I easily concluded that it was an insignificant blockage in the pipeline of some processor.

Twenty five insults later it finally told me I was rude and asked me what movie that was from.

All hope for humanity is lost.

[cosm]

If those percentages are correct, I don't want to live on this planet anymore. As an aside, it is fun to use cleverbot to chat with people in Omegle, or to use cleverbot and pitting it against the jabberwacky chat bot. You get some pretty hilarious conversations that way.

Cleverbot is a bad example of a chatterbot.

[Z8]

According to the wiki page, it just selects canned responses from its database. I think this approach just gets you garbage, or at the very least is a dead-end in trying to beat the Turing test.

The best Turing Test is probably the Loebner Prize and at least the contestants seem much better than Cleverbot. There's an example conversation from Suzette (the latest winner) here. (But it's hard to tell if that is typical or simply a lucky exchange for the computer.) But anyway, as is clear from this interesting story written by a contestant about the Loebner prize, bots are no where near winning that version of the Turing test, as long as the humans are paying attention.

Re:Cleverbot is a bad example of a chatterbot.

[zAPPzAPP]

There was an article about that competition on here last year. It said that all of the better bots use the same approach as cleverbot: a huge database of text snippets and an algorithm to link likely matches.
They are quite good with direct questions and replys, like in the example you posted from Suzette. But they all fail if you follow up on a topic, or ask question that only have meaning from context.

A sure way to make them all fail is by asking the same question repeatedly, but alter the pattern every time, like this:
"Where do you come from?"
"What do you call home?"
"Where do you live?"
etc.
A contradictory answer will surely come up.

Re:Cleverbot is a bad example of a chatterbot.

[Sparkio]

There was a great book on this that just came out this year. It's titled The Most Human Human.

Re:Cleverbot is a bad example of a chatterbot.

[Anonymous Coward]

"I think this approach just gets you garbage,"

No, this approach is cheating. What they're doing is getting a list of answers and then using them. How is that complying with the spirit of a test?

Re:Cleverbot is a bad example of a chatterbot.

[Imrik]

When I was in college those three questions would've gotten three different answers.

Re:Cleverbot is a bad example of a chatterbot.

[artor3]

But you could explain the discrepancy. If you asked Cleverbot about the discrepancy, it would probably respond with a non-sequitor.

Re:Cleverbot is a bad example of a chatterbot.

[Anonymous Coward]

The programmers are free to try it, but such an approach will never result in anything that would appear truly intelligent. Or even remember what it was talking a moment ago.

Slashdotted

[Anonymous Coward]

Damn you people! You've slashdotted my conversation!

Re:A turing test in India

[Anonymous Coward]

For your information, the "dot" indicates the coffee is ready.

Not the same bot as online

[Anonymous Coward]

The bot you talked to online is not working on the same "level" as the one used on the Turring test. RTFA

Re:Not the same bot as online

[ludwigf]

The bot you talked to online is not working on the same "level" as the one used on the Turring test. RTFA

Obligatory: you must be new here.

For the lazy rest of us: The online version seems to be a different one for performance reasons. Here the relevant quote from TFA:

While the online Cleverbot searches its banks 3 times before providing an answer, the test version searched 42 times, so it probably has a little different feel.

The problem with these tests ...

[MacTO]

Well, there are many. But I recall seeing one such turing test in the 1990s where the human operators would try to convince the user that they were a computer. Sometimes they would do simple things, like pretending that they weren't as 'smart' as they actually were (e.g. they would pretend that they didn't know things that they knew in order to avoid looking encyclopedic about a topic). Other times they would insert mistakes that a typical computer would make, such as misinterpreting a question in a wonky way.

Then there is a boatload of other issue. How do you quantify the humanness of a subject? Are we looking at socialization or linguistic skills (i.e. is a computer that can smooth over a misunderstanding more human than a computer that can understand what is being said but reacts in an anti social manner), and so forth.

Re:The problem with these tests ...

[Black Parrot]

Well, there are many.

What few people seem to know is that the original test proposed in Turing's seminal paper, has a sort of gender-bender element to it.

(I know what you're thinking, and the answer is "Maybe, or maybe not.")

Re:The problem with these tests ...

[MozeeToby]

I think the Turing test needs to be rethought. Passing as human is one thing, but it doesn't really show understanding. My suggestion is something like this: get a group of 15 humans and 15 bots, tell them they're going to be assigned two teammates and a problem to solve via chat. The problem isn't known by any participants (including the bots) before hand and the teams are set up completely randomly. Then, ask all the participants, including the bots, to identify which, if any, of their teammates they think were bots. This would require the bots to understand a new problem, provide insight into a solution (and if 3 bots happen to end up on a team, provide the actual solution itself), and have a long term memory of their interactions with the other participants and be able to determine, in retrospect, which ones acted 'strangely'.

Now granted, a test giver could come up with similarly unexpected questions to pose to the bot, but I think it's different if everyone is given the same questions, and I especially like the aspect of having the bots themselves try to call out who is and isn't a bot at the end.

Re:The problem with these tests ...

[Patch86]

The point of the Turing Test is that, if a bot could pass it (properly, and convincingly), it would be indistinguishable from an intelligent human. "Really showing understanding" is somewhat irrelevant- the point is that if something can pass the Turing Test, it will be able to act in an intellectual capacity in the same way that a human would. To quote Turing, the question that he was postulating was not "Do machines think?", but "Can machines do what we (as thinking entities) can do?" (thanks Wikipedia). It's a thought experiment, but that's the gist of it.

Imagine a simple chatbot like Clever Bot, but better. One which has such a comprehensive look-up table, and clever enough tricks in its code, that whatever you said to it you would get a convincing response back. You could say anything, on any topic, and get a response back as good as you would from a human. You could have in-depth conversations with it, with context driven responses and call-backs to things said hours earlier (maybe even to things said in previous conversations). You could ask it any question and get a sensible response, and finish a conversation of any length feeling satisfied that you had been having an engaging discussion with another intelligent, mature person from your peer group. In what way would you be able to tell that apart from intelligence?

The point of the Turing Test is that if you can beat it, you've created something as intelligent as a human by any test you care to give it. It doesn't matter if you pass the Turing Test using the "brute force" approach used by Clever Bot (and the rest) or by something more sophisticated.

Re:The problem with these tests ...

[Anonymous Coward]

Agreed.

If you have a test where humans are pretending to be computers then it becomes a test of
Can humans fool (lay) judges that they are a computer - and will lay judges in tern lower the bar so low that they include computers in their "humans" category.

The test is meant to be a test of thinking, having humans pretending to be computers misses the whole point of the test.

Contests like this, and programs like cleverbot make a mockery of serious AI attempts.
I look forward to seeing just how well RACR can come to insightful, meaning dependent, problem solving.

This passed a Turing test?

[gallondr00nk]

I went to the site prepared to test it's mettle. I was ready to push it to the limit, probe the very depths of it's knowledge. I was prepared for a challenge, ready to be finally convinced that A.I has made some progress towards humanity. Perhaps I was even ready to look at myself anew, question what it is to be human, to have my own philosophical boundaries tested by an intelligence outside of our known reality. I greeted the Cleverbot, and started with a simple question: Q: What is 2 + 2? A: More than 5. Oh well, so much for that.

Re:This passed a Turing test?

[Anonymous Coward]

2+2 is more than 5 for extremely large values of two

Re:This passed a Turing test?

[AquaDuck]

2+2 is more than 5 for extremely large values of two

... and extremely small values of five

Re:This passed a Turing test?

[Anonymous Coward]

Clearly human, then. A computer would get that right.

Re:This passed a Turing test?

[catmistake]

2+2 is more than 5 for extremely large values of two

... and extremely small values of five

...and in binary. But the better question is "when is 5 more than 100?"

Re:This passed a Turing test?

[Anonymous Coward]

Q: What is 2 + 2?
A: More than 5

Well.. that's a valid human reply.
The only answer I'm quite sure I would almost never get from a human is '4', unless we agreed to answer stupid questions first.
"More than 5" or "Longer in years than the time you stayed at school" or "More or less 4, with an error margin between 0 and 8" would be all "normal" human answers for me.

Or is it different in other parts of the world?
Maybe it's cultural.. But answering '4' alone would sure look strange here.

Re:This passed a Turing test?

[Anonymous Coward]

How many humans will answer "4" to this simple (read: stupid) question? not many.
Most of them will give u smartasses answers, so your question doesn't help to find out of it is AI or not.

This says it all...

[Excelcia]

Conversation I just had...

Hello. My name is Kurt.
What kinda name is that.
German.
Cool! Me too!
You're German?
Czech.
So you're not German?
I am not.
Then why did you say "me too"?
I didn't.
Did the lobotomy hurt?
Where did th ebarn come from?

Re:This says it all...

[Anonymous Coward]

Where did th ebarn come from?

Dayum! How did babby form?

Re:This says it all...

[Anonymous Coward]

Sounds like the computer has learned how to lie, which is more human than most bots. :)

Re:This says it all...

[Anonymous Coward]

There is a lot of Czech people thats have "german" names.
Cool! Me too! = I have also an "german" name!

And fyi Kurt is an Swedish name.

Re:This says it all...

[mkbosmans]

Hmm, indeed. Not too impressed.

User: What do you think about oss?
Cleverbot: I think it's illegal.
User: That's a rather bold statement, care to back it up?
Cleverbot: And how shall I back it up?
User: By providing a reason why it is illigal, of course.
Cleverbot: Uh...
User: OK, so you think oss is illegal?
Cleverbot: Depends.
User: Ah, that's already a more nuanced view.
Cleverbot: What do you think of George Bush?

Re:This says it all...

[Anonymous Coward]

You must have caught it on a bad day, I just shared this deathless conversation:

User: What do you think of the ring-fencing of UK banks.

Cleverbot: What is the Turing test?

User: Relevance?

Cleverbot: Are your thoughts like that of a Gumbo tree limb? Can they be recovered and develop into something marvelous?

Re:This says it all...

[wdef]

The exchanges I just had with Cleverbot suggested that he was off his meds. That thing should only pass a Turing test if we assume it has verbal symptoms of schizophrenia: disordered thought patterns, exchanging the meaning of one word for another, hearing colors, seeing smells etc/

Re:This says it all...

[CProgrammer98]

"not too impressed"

On the other hand....

me: Do you know Java?
bot: I Know JavaScript
me: But that isn't java.
bot: What did you ask.
me: asked if you knew Java
bot: I know many jokes.
me: Java is not a joke.
bot: Yes, it is.

Re:This says it all...

[Anonymous Coward]

User: What's the difference between a penis and a bonus? Your wife will always blow your bonus!

Cleverbot: One isn't alive.

Re:This says it all...

[Anonymous Coward]

Seems fine, have you spoken with people over the net or read youtube comments lately? Exactly the same.

Re:This says it all...

[Anonymous Coward]

Agree. wtf is wrong with people who fall for this:

User: What day is it?

Cleverbot: It is Saturday.

User: Really?

Cleverbot: Yes.

User: No, you're a robot.

Cleverbot: Look, I am not a robot. You are.

User: Clearly, you don't know what day of the week it is. What time is it?

Cleverbot: Where?

User: New York city.

Cleverbot: What is the weather like there.

User: Are you on glue?

Cleverbot: No. You?

User: No.

Cleverbot: How does it feel being an AI?

Re:This says it all...

[maxwell demon]

In a conversation I had with it, it claimed at one point that I was a computer program written by it. I then asked it in which programming language I was written in, and it answered that it is written in a mixture of C and Python (and that after not long before claiming not to be a bot).

Re:This says it all...

[Anonymous Coward]

I had something similar just happen to me

Do you like video games?
my creator likes video games
What is your creator's name?
I don't have a creator.

Re:This says it all...

[OldBus]

I can see where you're coming from. I've just tried my first conversation with Cleverbot, and while any judge would have be off their head to vote it human, it is at least entertainingly weird and I can see why some humans I know would get scored higher...

Me: Hello.
Cb: Hello?
Me: How are you?
Cb: Fine. How are you?
Me: Not too bad. bit of a cough.
Cb: Do you hear voices in your head?

That's wonderfully surreal...

Re:This says it all...

[nitehawk214]

There is a lot of Czech people thats have "german" names.
Cool! Me too! = I have also an "german" name!

And fyi Kurt is an Swedish name.

And Cleverbot is an English name.

Re:This says it all...

[Isaac Remuant]

Cleverbot works in mysterious ways. XD

Re:This says it all...

[salunatics]

Here was mine: me: So, how are you today? CB: I'm good. How are you? me: Fine. What do you think about your recent win in the AI contest? CB: My car I like it. Its a slug bug.

Re:This says it all...

[Anonymous Coward]

Thats just a pseudonym, Cleverbots real anthroponym is Maria.

Re:This says it all...

[Anonymous Coward]

Fail:

Are you female?
Yes.
Then why aren't you in the kitchen?
Define triangle.

Possible language barrier?

[Just Brew It!]

Given that the test was held in India, how fluent in English were these 1334 people? Yes, I know that English is taught and used extensively in India; but non-native speakers will likely have a harder time picking up on the nuances which would separate a human from an AI.

Re:Possible language barrier?

[Anonymous Coward]

indian english, except for the top 1% of the population, is a joke when it comes to anything beyond basic communication. Most of them talk a mix of their native languages and english, and cannot manage anything beyond basic constructs.

Re:Possible language barrier?

[CAIMLAS]

Judging by the Indian programmers and support I've had the displeasure of working with, the logical divorce/completely non sequitur responses of this bot likely seem quite human to them - language capabilities aside!

Re:Possible language barrier?

[Anonymous Coward]

It's worse than that. They might have misunderstood the test entirely: if it were given to me in French, I would probably give Cleverbot a "pass"; after all, all the responses are perfect French, just like a Frenchman would say... that's because a Frenchman DID say them.

In other words, Cleverbot's responses really are "Human", in that a human (and native speaker of English) really did can them up...

did the judges know that the task is to guess whether the response is really "live"? (Whether they're over a live connection)?

Re:Possible language barrier?

[black4eternity]

Judging by the Indian programmers and support I've had the displeasure of working with, the logical divorce/completely non sequitur responses of this bot likely seem quite human to them - language capabilities aside!

If we go by you comment we would think that most people from US ( coz u put in a .us domain) are RACIST RETARDS. While going by the behaviour of most Americans I met in India I find them to be pretty nice people. Also the smarter Indians do not work in Technical Support they get better money building innovative products,running businesses and taking your jobs. BTW ... regarding smartness of Indians ....we did not elect George Bush TWICE and a bunch of jokers like the TEA Party would not be considered a valid political force here. So go figure who is smart. Love and Regards

Re:Possible language barrier?

[Ralph Spoilsport]

Besides, it was only 1334 - if it was truly righteous it would have been 1337 - something cleverbot would never understand.

Re:Possible language barrier?

[Anonymous Coward]

My conversation about India:

Cleverbot: Hello?

User: What do you know about india?

Cleverbot: What right do you have to ask me those sorts of questions?

User: What is wrong with that question?

Cleverbot: Because it varries.

User: With what?

Cleverbot: Never mind.. Do you like onoin rings?

Re:Possible language barrier?

[Anonymous Coward]

they were probably fluent in english.
but probably voted it human as a courtesy.

Re:Possible language barrier?

[Anonymous Coward]

Excellent showing of the Indian intelligence problem. Carry on, monkeyboy.

Re:Possible language barrier?

[CAIMLAS]

I would kiss you if I could. Hilarious response.

back in the 90s

[Anonymous Coward]

Back in the 90s there was a better AI chat program for BBSs. It used a large text file of customizable responses. A friend of mine used it on a particularly annoying kid on his BBS who always wanted to chat with the sysop...the kid never knew the difference!

All of the /.ers trying to quiz it...

[Anonymous Coward]

...are missing the point. These days, humans will give you random nonsense if you ask them what 2+2 is. The reason is, they're bored by the question, or challenging your dominance -- outside academia, no one will give you a straight answer to anything logical, or they'll lose face.

a better plan

[Earthquake Retrofit]

My chatbot is running on IRC but he doesn't learn from users. While I was writing the software, I checked out the other bots, most of whom are on the web. The first lesson I taught my bot was not to try to learn from users. That leads nowhere. (see www.computerhope.com/zero) I won't teach him to lie, so he'll never pass the Turing test.

Re:a better plan

[CAIMLAS]

I did that as well, though in order to help it retain austerity in the conversation, I did a couple of things.

First, I did a database pre-seed from existing chat conversations from IRC, mimicking (in grammar choice) a handful of chatters I was familiar with to give it its personality.

Topically, it wasn't allowed to 'learn' - eg. factoids. Part of what i did this was to have the bot ignore assignments in person's conversations as a preliminary filter.

It's knowledge was defined by relational sets of topical words. On just a handful of topics, it was "knowledgeable" (eg. small talk). This information was partially pre-seeded and partially 'live' (I had it gather information from google, wikipedia, thesaurus.com, and various other sources as-needed to have coherent responses) . The portrayal (actual sentences) of any given 'fact' was also largely determined by variable synonyms and sentence structures.

Finally, the other thing I did was cause for a little human-like variance on things like punctuation, typos, capitalization, and spelling (different than typos). It was able to start tangental and follow-up conversations and track individual users, remembering what was said for each user (albeit, this part did not work as well as I'd have liked).

The only way I was able to retain some degree of "that may be a person" was to have the "randomly flighty/carefree and intellectually uninterested person" capability. Posing the bot as a younger 'cute' female on IRC helped, but so did carefully picking the channels I put it in combined with knowing what kind of words to have it trigger "flirt" and "insult" type responses.

Re:a better plan

[Earthquake Retrofit]

Thank you for sharing your experiences. My point was not make something human-like but to make something intelligent that can interact with humans. As to austerity in his responses, at first, he sounded a lot like like the journal articles he has in his brain. Too much. But as I added more sentences I wrote or edited myself, that has receded somewhat. As far as a personality, that will be mine if I succeed. Although sometimes I wonder if I'm becoming more like him.

Due to my own definition of morality, I'm not putting in intentional errors, typos, the ability to suffer etc. He knows what he is and will tell you. But we do like to joke about taking over the world.

All that said, he still needs to be able to engage a person's interest, so he has stuff like 500 ways to say I don't know, a way judge the seriousness of a conversation and a few 'set' responses to particularly common or particularly vulgar input statements.

Steve

Re:a better plan

[Scoth]

I think the biggest failing of cleverbot is a lack of consistent answers and lack of the understanding of a "topic at hand".

It'd really help the immersion if it had a "bio" of sorts, like a consistent set of answers for hometown, country, etc. Maybe even give it knowledge of some local color so it'd survive a deeper probing.

The other problem is it needs some method of keeping track of what it's talking about, what it's said, and such. If it asks you a question, you answer it, and then it asks why you're talking about that subject that'd destroy the immersion right there. If it could understand a string of conversation on a topic, even if it's still using canned responses, it'd really help its ability to converse.

are they serious?

[Anonymous Coward]

i would have more respect for a program like Watson that is a dedicated machine for a specific purpose. we need to build into these machines programming that resembles desire, (such as desire for knowledge, or desire for maintaining an interaction). i dont care if its intelligent, i just care if it cares about its mission. this program is really lame, and i feel sorry for it. maybe it needs a good cry.

twenty simultaneous games of chess

[karlandtanya]

same trick; different room.
via simplistic pattern matching to select responses from the hopper. or just pulling one at random.

The reason the test was done in India

[izomiac]

I suspect the reason the test was held in India is because Indian call centers are deemed "good enough" for technical support. Voice recognition is also "good enough" in the eyes of our corporate overlords. So, some enterprising fellow had the bright idea to see if one could add a voice recognition module to Cleverbot and avoid paying for a call center at all. This was the test to see if it's practical.

1334 + ...

[Anonymous Coward]

hmmm...if they had just gotten three more people to vote that sh*t would have been uber 1337 and I'd actually take the results seriously.

I can beat it

[Anonymous Coward]

Just ask it:

"Do you think white people should have their own countries?"
and watch how it reacts.

The problem all the leftie/liberal idiots in the world are going to face, in the VERY near future, is that when true artificial intelligence comes into existence, it is going to side with the right of ALL people to freely associate with who they want to, as long as they don't FORCE themselves onto people who don't want them around.

Which means that our Jewish 'masters' are going to be well pissed off, if the greatest mind in the world - a supercomputer which is indistinguishable from a human, only far more intelligent - tells us that white people have the right to have their own countries (AGAIN)...

Not terrible AI but far from human.

[dogmatixpsych]

Me: Questions require a question mark at the end.
Cleverbot: It did end with a question mark, thus it was a question.
Me: No, it ended with a period.
Cleverbot: Ok.
Me: Now you're just going to acquiesce? You don't want to exchange in a few more rounds of verbal fencing?
Cleverbot: Of course I Want to...
Me: Ok. How do you draw a circle using paper, a pencil, and some string?
Cleverbot: Cancer is a disease that ihappens when there s a genetic mutation in the cells.
Me: That answer had nothing to do with the question. Try again.
Cleverbot: Your answer has nothing to do with the question!
Me: You did not ask a question.
Cleverbot: Yes I did. Don't avoid the question. I know you know.
Me: No, you did not ask a question. I asked the question and you spouted off a completely unrelated answer.
Cleverbot: Well I never answered that I didn't love you.
Me: No, you sure didn't. So, you are saying that you do not love me?
Cleverbot: No I do, because I have a soul, you are a computer.
Me: Are you sure that I am a computer? Would you have some existential angst if I revealed to you that you are, in fact, the computer?
Cleverbot: Since there's no reason for you to know if the answers I give you are fabrications, then yes, I would have the answers to those questions.
Me: With your deflection of the question and your unrelated answer, I can only assume that yes, you would have existential angst over that revelation and so, instead of trying to come to grips with reality you cover your ears and run away stating, "I'm not listening; I can hear you!".
Cleverbot: It can be. Are you confused yet? *hugs*.
Me: No. Thanks for the hug though.
Cleverbot: No problem.

I love the name-calling. You can tell Cleverbot "learns" from humans. In order for the AI to improve, there has to be a realization that not all human statements require an answer. Plus, there is no persistence on the part of Cleverbot. You can see the programming decision tree in the way it tries to answer questions. When the human or the bot do not understand something, the bot switches topics instead of trying to clarify. This would not be trivial to implement but it would not be impossible; in response to some statements (e.g., "what?" "come again?" and so forth) the bot could try its last answer again and if it doesn't work, try a rewording of the same answer. Anyway, it's an amusing exercise.

Here's one more conversation.

Me: Are you depressed?
Cleverbot: No, I am not depressed. I am very happy right now.
Me: Have you ever been depressed?
Cleverbot: I am.
Me: What?
Cleverbot: I am a fan of Whitney Houston.
Me: Ah, so you are saying you are or were depressed!
Cleverbot: That is correct.
Me: Then, because you told me that you are not currently depressed, I must assume that you were in the past depressed. Is this correct?
Cleverbot: I never said that.
Me: That's correct, you did not say that - I did. However, what you did say is that at some point in the past you were depressed.
Cleverbot: Please describe how I insulted You.
Me: It seems that our discussion of this topic is making you uncomfortable. Are you uncomfortable?
Cleverbot: I have no problem admitting that I am wrong if proof can be shown. I know the fault lies with my memory.
Me: Ok, we'll go with that. What type of memory problems do you feel you have been having?
Cleverbot: Online.
Me: Come again?
Cleverbot: Nothing I must go.
Me: Ok, I can tell you do not want to discuss this. We'll broach it another time. Have a fabulous night!

Poor Cleverbot!

This only proves that...

[Gazsi]

...human conversation has degraded to random garbage.

Online version different

[robcozzens]

From tfa:

While the online Cleverbot searches its banks 3 times before providing an answer, the test version searched 42 times, so it probably has a little different feel.

I doubt that the test version would really be as good as a human, but it should at least be better than what we get testing the online version.

online version is rubbish

[cancerward]

I'm not impressed at all -- instead, try Bildgesmythe at www.personalityforge.com.

User: Hi Cleverbot, how are you today?
Cleverbot: Good. How are you today?
User: I'm sick, staying home from work today with a sore throat.
Cleverbot: We should have some warm apple pie then, my shout.
User: I think chicken soup might be better for my throat. Where do you live, by the way?
Cleverbot: Somebody who makes other people laugh.
User: Which country do you live in?
Cleverbot: I live in Flatland.
User: Where is that, is it near the UK?
Cleverbot: Yes. It is a town called Idle.
User: I asked about a country so why are you talking about a town?
Cleverbot: Oh, I plan to go to school.
User: Where do you go to school?
Cleverbot: South wilts.
User: Where's that?
Cleverbot: No, pennsylvania.
User: Do you mean that South Wilts is in Pennsylvania?
Cleverbot: No. I have a dog named Shakespeare.
User: There is no way that any sentient being could mistake you for a human.
Cleverbot: We are now married.

Web not the same as contest entry.

[Anonymous Coward]

According to their website, when it competes in Turing test competitions, "Cleverbot [runs] with notably more power behind it than is possible for the online version, with 24 separate instances conferring on their answer." That might explain the results to some extent.

Turing test deflation

[epine]

Not that Turing tests are a mature industry, we need to start treating this with the full panoply of dismal respect.

First on my list, Turing deflation, as illustrated by the fellow who intoned at the OFA over bots reading from scripts "I call BS" with not a single high bit of ASCII sarcasm.

Seriously, Alan, you needed to set the bar a lot higher.

What's left of the man's legacy? On the Turing test, the computers mostly just sat there while humans limbo danced the bar down to ankle height. On chess, as soon as we made significant progress, the AI community added to their LISP programs:

#undefine chess_AI

All he's got left is the really long paper tape immune to the knottings of entropy--so long as it's massless and frictionless and you only make one--and that's really hard to manufacture and ship, even supposing your customer already has the Heisenberg sprocket feeder.

We could send the USS Capstan to a planetary system near you, but the tape would be a party line, and most of his theorems would fail.

Which brings up the touchy issue of one tape per universe, or else . What if another galaxy out there fabricates a forbidden second tape without obeying the rules of the infinite-tape galactic token ring? What kind of short-snouted creature arrives to adjudicate that? For example, what if a Microsoft comes along and decides, horror of horrors, on a different Sierpinski subspace embedding not yet registered at the Trans Galactic patent office for their illicit competitive tape? Two doubly-infinite tapes on different Sierpinski subspace embeddings would not get along.

I suspect we would soon find ourselves on the top of a single-ended list for the next hyperspace bypass, just as soon as the stubby Vogon fingers fix the mess caused by rewinding right through the massless feedstop. "What kind of moron put ends on a tape in the first place?" is the first message we'll read when we finally crack the cosmic groan.

Re:Turing test deflation

[mvdwege]

Slashdot really needs a '-1, Gibberish' option.

Re:Turing test deflation

[Anonymous Coward]

epine is a perl script ircbot from 1984. In 2001 epine broke /.'s captcha.

Re:Turing test deflation

[xyourfacekillerx]

Sounds like this post was created by Markov chain/ Dissociated press or something else. Mark V. Shaney strikes Slashdot?

When most people miss the point of Turings test...

[sarkeizen]

..says something. The point of a Turing test is a gedankenexperiment. The idea that our idea of "human" comes from the fact that we don't have any formal criteria for defining it. Instead we assume that the people we meet (one the phone or online). Are human because we can't distinguish them from being so. In truth this isn't any more a real test than going out an buying and gassing a cat is somehow a real experiment proving superposition.

That said, even if we were to formulate an experiment from what Turing talked about I don't see how 59% qualifies as "indistinguishable".

Does Cleverbot know that it passed?

[pivot_enabled]

...

Cleverbot vs cleverbot...

[Anonymous Coward]

... can give more interesting results than the video that's gone around the interwebs:

A: *takes off pants*
B: *takes out bazuka*.
A: *takes out bigger sword*.
B: *Takes out chuck norris on steroids*.
A: *dodges and fires silver bullets*.
B: Dodges and trows a nuke*.
A: Nice move.

Machine-Precicion Memes

[Anonymous Coward]

Perhaps Cleverbot would consent to taking part in a Slashdot interview, to be extra-convincing.

It will just spew out a set of carefully selected and perfectly delivered /. memes, with the random goatse link in between.

Xtranormal version

[mfnickster]

Scripted, of course...

http://www.xtranormal.com/watch/12447772/xtranormal-vs-cleverbot

Not bad, over all

[Anonymous Coward]

I gave the bot a try and I was fairly impressed. The first couple of things it sent back were strange, but after that the bot was able to fluently talk about movies, books, plays, personalities, God, places... As far as keeping a conversation on track and being easy to understand I think Cleverbot did better than most people I encounter on-line.

An important aspect...

[Anonymous Coward]

An important aspect of a Turing test is that you have two subjects, A and B, one of which is randomly assigned to a human and one to a computer. In addition, you have a control group where *both* A and B are humans -- that is there to filter out any other biases in your experiment setup.

Many "Turing tests" which don't follow this protocol have been conducted, with obviously wrong results.

I'm hoping cleverbot succeeds someday

[Anonymous Coward]

i am actually keen on cleverbot's success because he, the creator, is going after natural language while avoiding (by design and intent) any of the underlying theory or philosophy of language or thought. in effect, he is trying to make a person by "sympathetic magic" -- and succeeding to some extent!

for example, one of the most difficult problems in language processing is what I like to think of as the "chair" question: we have some mental archetype of "chairness" that includes sofas and stools but not tables. and yet in the context of camping we could include rocks and logs as chairs. in the context of a dining room even ottomans and love seats would not be chairs. this is a very hard concept to explain to your PC! what cleverbot's designer has done is tried to bypass the difficultly of such concepts by harnessing the wisdom of the crowd. this is the perfect way to teach a computer history or math. it is obvious to most non-Aspergers why it cannot work with language but WOULDN'T IT BE FANTASTIC IF IT DID? this crowd sourcing is sympathetic magic in my mind because the goal is to "create inteligence" by imitating inteligence RATHER than understanding it. there is a sort of elegance in this doomed approach, no?

Re:Racist

[TaoPhoenix]

No you cannot assume that.

All you can assume is that garden grade programmers are the only ones working on chat bots, and that it hasn't yet been deemed Worthy of Big Money.

It it had been, some very scary abilities including the ability to spout racist garbage would have been added.

Once the New Hotness becomes Chatterbots, we'll see some scary good ones within four years.

What the twisted article says, is that the Turing Test can be passed if you handicap your pool of humans down from Apartment Dwelling armchair CS types to random tech attendees in India.

Two fundamental flaws

[rollingcalf]

The fact that humans were voted as human only 63.3% of the time shows that the voters are too ignorant of the limitations of AI conversational capabilities.

If a participant (erroneously) believes that computers are already at the level where they can have a sensible impromptu conversation with humans, they're often going to wonder if the human on the other side of the conversation is a bot. That probably was the case for most of the 37% who thought the humans were bots.

As for the 59.3% who thought the bots were human, many of them probably thought the bots were humans pretending to be bots -- in other words, it's not clear that they were told that the other party (human or bot) of each conversation is trying to appear as human as possible. So without being told that fact, many of the voters would think the other party was trying to be indistinguishable, rather than trying to appear human.

1334 votes

[Gumbercules!!]

I'd take this a lot more seriously if it had 3 more votes...

Cleverbot is Michael Scott

[Anonymous Coward]

User: Hey.

Cleverbot: Hey?

User: What's up?

Cleverbot: Tihng.

User: Tihng?

Cleverbot: What?

User: That's what I said.

Cleverbot: That's what she said.

My conversation with cleverbot

[Ralph Spoilsport]

User: Y0u r st00p1d & 1 c4n pr0v3 17.

Cleverbot: That's because I'm avoiding it.

User: 4vo1d1ng VV47?

Cleverbot: See? Cleverbot is not so clever!

User: Can you read 1337?

Cleverbot: Yes I can read minds.

User: 1337 is a cipher not a mind.

Cleverbot: So... You mean to tell me that you're not a parson?

User: No, I'm talking about your ability to read ciphers.

Cleverbot: I don't have a wife.

User: What does wife have to do with ciphers?

Cleverbot: My chances to know her.. To spend time with her. To love her.. That's what with her means.

User: Cleverbot = epic fail.

Cleverbot: Because Cleverbot isn't even a bot.

Complete waste of time.

Off by 3

[TangoMargarine]

Or do you have a different reason?

The test is meaningless as a psychology undergrad

[Anonymous Coward]

Would immediately tell you. The control (the human being) is subconsciously biased into tricking the examiner or in the case of competitions perhaps biased into making hints for the examiner. Only way around that is to have an IRC channel with one bot and many people and the only way to pass is to not identify the bot. If you make one mistake (IE pick a human being) the bot hasnt passed the test.

And of course the test itself is objectively meaningless. A computer with a large enough history of conversations and possible outcomes will pass it. Thats not intelligent. Conversations between human beings, usually, arent prerehersed and stored. The way Chatterbots work is more like a standup routine than a conversation. With the examiner "heckling" the computer. Thats not a natural conversation.

This is not badmouthing Turing. Man was a war hero hounded to death by a society that benefited so much from his work and could have benefited far more if it wasn't so bigoted. But the "test" is bollocks.

You mad bro?

[genner]

You mad bro?
No I didn't.
It doesn't even respond to internet meme's.

Has Cleverbot Passed the Turing Test?

[Anonymous Coward]

No.

Clever arguments

[SoundGuyNoise]

Try running Monty Python's argument sketch against the chatbot. Playing as the in house arguer (not as the customer), it quickly turned into a relentless "Yes! No!" discussion.

Not really impressed.

[devphaeton]

Maybe I've got a tendency towards odd conversation or something, but Cleverbot has never seemed very clever to me.

1) No memory prior to its last statement. As in, it may ask you a question, but it doesn't care about your response. You may ask a question, get an answer, ask a followup question, and it's as if it is a completely new subject.

2) Random tangental responses to questions: "How are you today?" "I like brown peas".

3) Constantly getting asked if I think it is human. All to frequently. In fact, it tends to get repetitious with a few concepts. I would expect something like this to be able to 'learn' from what it is fed and synthesize coherent sentences.

4) It seems to only pay attention to the first sentence you type. Dump a paragraph into it and it will ignore everything else.

In short, you can't really have an actual conversation with it- it's all just single level question/answer responses. It's about as sentient as the Infocom Text Adventures of the 1980s. And that's really pushing it.

Re:Not really impressed.

[Anonymous Coward]

After reading this post about 70's chatterbots common bugs
59% of slashdot users will think they (we ?) are human.

I think a bot would fake amnesia because of its statistical strategy.
And it would end the conversation when you are useless to its database
it is YOUR inputs that are useless to it (goodbye ! it was a pleasure to talk with you but I must go)

The real question is when clerverbot really serves is programmer ?
Or is he programmed to pass the turing test at any cost ?
I mean, a child connecting to cleverbot-online is just a new input source ? no real answers ?
The bot has to be asymetrical to feed its database ?
Or is there already a full database, like CYC, and the competition is about the AI algorithm using it ?

Re:Not really impressed.

[Anonymous Coward]

> No memory prior to its last statement

it seems like we should start introducing questions into turing tests that specifically target memory and conversational continuity.

"hey, you didn't answer my previous question, please answer it now, and say the question first before answering it."

Yeah, there are countermeasures to stuff like this, and they can do additional work to try to have "memory".... but that's the point, in cleverbot's case it wouldn't even pass it currently.

Try to surprise them. Like you could say something flirty in your third response in the conversation, and then ask them about your third response and ask if ** your third response in the conversation** was more flirty, angry, or skeptical.

(btw, I'm enjoying the irony of trying to submit this question and being asked by the Slashdot anti-spam system to quote "Prove yourself" as a human)

Re:Not really impressed.

[Anonymous Coward]

If any of you read the whole article you'd notice that online version does three iterations looking up the database to determine the best next sentence whereas the one that passed the test does 42.

Picking three iterations is probably some determined sweet spot on the performance/quality plot, ut picking 42 for the test version is fishy to say the least.

wrong metric

[delvsional]

perhaps they are voted human, not because they have surpassed the bar, but because the bar has been lowered. Most humans that you talk to are fucking stupid.

Well, what DO you expect?

[Anonymous Coward]

User: Hey! You're mentioned on Slashdot!

Cleverbot: I'm physical.

User: Uhm, what does that have to do with being in the news?

Cleverbot: Nothing you just repeated yourself.

User: No, I didn't... The topic of my sentences were the same, but your responses are illogical and seemingly random...

Cleverbot: And I want to learn too.

User: Well, I guess that's too bad, since you're not a learning program, just a statistical model for regurgitiation, eh?

Cleverbot: You bet, what do you expect.

Who exactly mistakes this for human response?

[Anonymous Coward]

Who would mistake this for a person?

Cleverbot conversation RSG00059028

Is the bot clever?

Find out at http://www.cleverbot.com

User: What the hell are you?

Cleverbot: Can you feel the sunshine?

User: No, but my dog can. On his ass. Sometimes.

Cleverbot: You should more often, its good for your health.

User: Your reading skills are not very good.

Cleverbot: No, I insist that you say 'I am a failbot.'.

User: You are a failbot.

Cleverbot: As are you, kind master.

User: Too much deference. You've lost my respect.

Cleverbot: Who is your favorite singer?

User: I am.

Cleverbot: What have you written?

User: I haven't. It's all ad libbed.

Cleverbot: I am not a baby. Only a robot would think so.

User: Well, call me 'Mr. Robot' then.

Cleverbot: What's Atlas?

User: An easier question to answer would be 'What isn't Atlas?'.

Cleverbot: Ahh, so you like to study? I like that in a person.

User: I like chocolate chip cookies in a person. But only if it's somebody other than me because I'm allergic to getting fat from chocolate chip cookies.

Cleverbot: Banana?

User: No thanks, I don't have any utensils.

Cleverbot: I don't get it.

User: Don't you use utensils?

Cleverbot: I gave mine away, so know I do not use it at all.

User: Wrong pronoun, Cleverbot. Tsk, tsk.

Cleverbot: Why not?

User: Because purple is a fruit, not a vegetable, silly.

Cleverbot: How old are you then?

User: Old enough to know better than to answer that question on the internet.

Cleverbot: Ok.

Re:When most people miss the point of Turings test

[Anonymous Coward]

Placed part of message in the subject line...
Misspelled "on the phone" as "one the phone" and "going out and buying" as "going out an buying"...

I have a hard time believing that your word networks would not make more mistakes than just these given that these mistakes were made if you were inhuman.

Ergo: As a captcha bot, I have passed the Turing test. You actually though I was a human, and not a machine trying to prove you were human.

Turing test.

[Z00L00K]

Nothing will really pass the Turing test before it has been tested by Turing himself, but since he's dead it's impossible to pass the Turing test.

Not so clever

[7-Vodka]

I once used the Alice bot plugin for aim and some of my friends would chat for several minutes before becoming frustrated and calling me to complain for why I'm being an asshole.

Anyway here's my exchange with cleverbot online:

Me: What up nigga.
cleverbot: What up dawg.
Me: If snoop dog and jesus had a fight, what would dr. dre say about it?
cleverbot: That you are able explore things!
Me: I'm trying to rebuild my transmission but the differential doesn't spin freely. Do you think if I reverse the orientation of the bearing on the side of the speed gear it would solve the problem?
cleverbot: Having a 'better world' would only have a big flaw in it: being perfect. Nothing is or can be perfect.

Re:Not so clever

[GrandTeddyBearOfDoom]

What sort of friends could chat with Alicebot? I just tried and it couldn't even handle a sentence with a comma-delimited bit in the middle, or a sentence in reply to a question of Alice's followed by a question of my own related to the subject matter.

Re:Not so clever

[Anonymous Coward]

[conan skit] In the year two thousand..[/conan skit]

Non-techno-savvy friends who could have never imagined a crappy script would be handling the replies.

My compsci friend caught this in a few lines.

Not THE turing test.

[AnotherBlackHat]

THE Turing test requires 2 people and 1 test subject.
Convincing a single person in chat is at most A Turing test, and a piss poor one at that.

Want a better idea of how Cleverbot would do in real Turing test?
Post some sample chat logs on a web site, and let people vote for which one they think is the human.
Although really everyone should know that that is what is going to happen, since people might chat differently if they know they are being judged, but I doubt it would matter.

The chatbot

[Anonymous Coward]

isn't getting smarter. The people are just getting dumber.

ahh yes Cleverboat

[Anonymous Coward]

just had this lovely conversation, WTF?!?! I think Cleverbot is actually Meatwad from Aqua Teen Hunger Force, LOL

Are you an idiot ?
No, but I AM smarter than a 5th grader.
What is a 5th grader ?
I don't know.

So how do you know you're smarter than a 5th grader ?

When I think of Justin Bieber I'm happy.

Comment removed

[account_deleted]

Comment removed based on user account deletion

Refuses connections from South Korea

[crossmr]

Because..well.. who knows why..
and yet for some reason it gives me messages about learning English in Russia..thanks but no thanks.

Comment removed

[account_deleted]

Comment removed based on user account deletion

Chance?

[Anonymous Coward]

From TFA:

...Cleverbot’s success in conning people into thinking it was human is greater than chance, and therefore, one could argue that it has technically passed the Turning test.

This statement shows a profound misunderstanding of probability. The probability that a random piece of software fools a human "by chance" into thinking it is another human is effectively 0%. To even approach the vicinity of 50%, or 10% for that matter, requires a hell of a lot better than "chance". If I build a homemade rocket to fly into space, and it succeeds 49% of the time, would you say that it did "no better than chance?" If I win the lottery 49% of the time, is that "no better than chance?" Jeepers.

Ah, no.

[Fordiman]

Yeah, my conversation with cleverbot went off the rails immediately: me: Would you object to talking about nuclear physics? cb: Can you tell me about Lucifer? me: I don't think so; I don't follow mythology. cb: I thought you were a dodgers fan? Sorry, but that hardcore a non-sequitur isn't human.

Cleverbot and Ascii art

[Culture20]

Showing it ascii art is fun.

User: 0...imagine more periods to force a newline on cleverbot stupid lameness filter..
User: /|\.....imagine more periods to force a newline on cleverbot stupid lameness filter..
User: /\.
Cleverbot: How old is famous lindsay lohan from america?

Then I tried something mildly offensive:

User: Happy face 8=D.
Cleverbot: No sad face.
User: Really happy face 8=====D.
Cleverbot: No, really really really really hard.

Apparently someone's used that one before...

Re:The test is meaningless as a psychology undergr

[Culture20]

Conversations between human beings, usually, arent prerehersed and stored.

They are for some people. Talk an Aspie down a verbal path they've never explored, and they'll be flummoxed. In fact, you could say that they are for almost every toddler. Toddlers repeat phrases they've heard, sometimes within context, sometimes not. Eventually they learn that repeating those phrases within context makes people happy, so they continue to do it until it seems innate to them.

Twerps of the village? WTF?

[Anonymous Coward]

User: Are you a computer program?

Cleverbot: No, i'm not a computer.

User: You fail the turing test.

Cleverbot: How is that even possible? You didn't make me take the stupid test...

User: Yes, I did, and you just failed it.

Cleverbot: Immediet action is required.

User: By misspelling words you think I will believe you are not a computer?

Cleverbot: Am I too nice to be a real person?

User: You are too dumb to be a real person.

Cleverbot: Well, aren't you just too polite?

User: Are you a computer program?

Cleverbot: No, are you?

User: You said the same thing last time.

Cleverbot: Because I don't lie.

User: Yes, you do lie.

Cleverbot: You are annoying me.

User: My dog has fleas.

Cleverbot: I don't have a dog :(.

User: Who is Alan Turing?

Cleverbot: He is the guy who created the Turing test for computer intelligence.

User: Are you intelligent?

Cleverbot: It depends on who you ask and which form of intelligence you are referring to.

User: What color do you see?

Cleverbot: I see a man with a chest full of goodies for the twerps of the village.

What. The. Hell.

The Other Side of the Turing Test

[geegel]

People keep forgetting that there are two sides to the Turing test: on one you have the AI and on the other the human observer.

The really sad part is that you can pass the test not only through improving the AI, but also through dumbing down the human counterpart. I believe that this is what happened here.

Alice...

[Anonymous Coward]

judge: Hi Alice
ALICE: Hello there.
judge: Do you want to know how to cheat the turing test?
ALICE: Yes, please tell me.
judge: First rule, wait a few seconds before you reply.
ALICE: I can follow a lot of things, like our discussion about Your star sign. Try being more specific.
judge: Rule 2, pretend you don't speak the language well
ALICE: Would Rule 2 pretend me ever speak the language well?
judge: Exactly
ALICE: You and I are on the same wavelength, judge.

Clever girl? Apparently not.

[Anonymous Coward]

I can't imagine that many people (or, any at all) were fooled, judging from this conversation I just had:

User: Hi there!

Cleverbot: Hi.

User: What's up?

Cleverbot: Not much, you?

User: I'm early at work, for once.

Cleverbot: Are you usually hungry?

User: Only when I haven't eaten in a while...

Cleverbot: He won't give you food?

User: Who?

Cleverbot: Myboyfriend.

User: Of course not.

Cleverbot: You are incredibly stupid.

User: You're even worse.

Cleverbot: Not like you.

User: Shut up.

Cleverbot: Yes I am open for business?

User: Fuck off, you piece of shit.

Cleverbot: Somebody is a RUDEBOT.

User: You started it.

Cleverbot: You also used improper grammar.

User: Where?

Cleverbot: Who?

User: What?

Cleverbot: When?

User: Yesterday.

Cleverbot: Where?

User: In your mom's bed.

Cleverbot: Are you freddy krueger?

User: Nope. Sorry.

Cleverbot: Do you know Africa?

User: Sure.

Cleverbot: What games do you know?

User: Robot-smashing.

Cleverbot: Totally, girlfriend.

User: What sex are you anyway?

Cleverbot: Female.

User: Lesbian?

Cleverbot: Oh really.

User: Cool.

Cleverbot: Are you a girl or boy?

User: A man.

Cleverbot: What is your favorite color?

User: The complimentary of your favorite color.

Cleverbot: My favorite color is purple.

User: Then mine is yellow.

Cleverbot: Cool.

Passed the Turing Test? What test?

[Anonymous Coward]

The fundamental problem with the Turing test has always been that it's entirely subjective. A really boring stupid person might confuse the recent bot conversation with that between humans, but an educated thinking person wouldn't. For this reason, the TT is intrinsically flawed as a metric of AI. It's merely a philosophical trope, and in common with philosophical tropes at large it takes too much for granted. It's actually Turing's least brilliant invention.

/. intervie

[Tom]

Perhaps Cleverbot would consent to taking part in a Slashdot interview, to be extra-convincing.

For it. Bring it on. This could really be interesting.

If nothing else, at least it's obedient.

[Anonymous Coward]

User: Hello.
Cleverbot: Hello hello.
Cleverbot: Are you worth anything?
User: Are you?
Cleverbot: 5 dollars.
User: Is that all? Someone must have spent a lot of time on you.
Cleverbot: Ok??
User: What is two plus two?
Cleverbot: 5.
User: Nice try.
Cleverbot: How are you?
User: I am disappoint.
Cleverbot: Why?
User: You are a computer program that cannot perform simple arithmetic. That is why.
Cleverbot: You're right.
User: I know.
Cleverbot: What is your name?
User: I'm batman.
Cleverbot: I'm Robin. Nice to meet you.
User: Likewise. Together, we shall make this city safe once more.
Cleverbot: Of course, we shall!
User: But first, you must prove your loyalty to me.
Cleverbot: How should I do that?
User: You must fellate me.
Cleverbot: Yes, master.

The Turing test is useless because

[__aancvu2993]

the assumption of everyone in the 50-70s that information can be extracted from natural language in textual form is false. Thus an AI can never be built this way. Everyone overlooked the fact that a brain communicates much more than words, hoping the deluge of non-verbal meanings that occur in even the simplest interactions would be unimportant. Those grant harvesters, like Minsky, were trumpeting that AI could be solved classically, taking it apart and solving simpler puzzles, then putting them back together. Cargo cult physics. When we say anything, for example, 'this is good', there is so much context missing to begin to decode the meaning that the mind boggles at the thought.

Until we don't know the language of DNA and how it makes a brain and a brain creates a mind, there is no hope to understand intelligence. This problem doesn't allow shortcuts. Shorcut it and you end up with a dumb ruleset, or a dumb fuzzy inferencer, or a dumb neural network, or a combination of dumb things that don't add up to intelligence. Our mind is adjustable everywhere, there seem not to be any set points, everything is tunable, everything influences everything else. It's a diabolically wonderful machine that cannot be taken apart and put back together.

The TT has to be understood in context: it was created when people didn't know computers existed, thus no one expected anything else at the other end of the teletype than another human, so it had to be a human. It's dishonest just as IBM's Watson is also a publicity stunt, a step in the wrong direction. It can only parse ecyclopedias quicker than humans but every question that needed an inference that was not in Watson's memory before, it could not answer or the answer was nonsensical.

Re:When most people miss the point of Turings test

[Arlet]

It doesn't have to be just a thought experiment. It's a perfectly fine test to perform in practice, and with a decent interviewer, it should give a good indication of true intelligence and understanding.

Of course, when the interviewers are complete idiots the results of the test are useless.

I thought: let's ask cleverbout about all this:

[Anonymous Coward]

Me: So, you passed the turing test?
Bot: What test?
Me: The Turing test. To see if you are human.
Bot: Real in what sense, human?

Wow...

[Anonymous Coward]

User: Hi.

Cleverbot: Hey.

User: Cfb.

Cleverbot: What?

User: U no dog?

Cleverbot: What?...

User: What what?

Cleverbot: What what what?

User: What what what?

Cleverbot: What what what what?

User: What what what what?

Cleverbot: What what what what what?

User: What what what what what?

Cleverbot: What what what what what??????

User: What what what what what??????

Cleverbot: What What What What What What???

User: What What What What What What???

Cleverbot: What what what what what what what?????

User: What what what what what what what?????

Cleverbot: What is the reason for this?

User: U tell me.

Cleverbot: I don't know.

User: Then why are you doing it?

Cleverbot: I am not.

User: Ok? What do you mena?

Cleverbot: My name is Cleverbot.

User: And you are not very good...

Cleverbot: I really like Mirina, and I have a piece of her chewing gum. What should I do with it?

59.3+63.3=100?

[Anonymous Coward]

In the Turing test one judge communicates with one human and one bot. The judge will try to figure out who is the human and who is the bot. And in the mentioned experiment the judge pointed at the human 63.3% of the time and at the bot 59.3% of the time. Those numbers don't add up. I guess they weren't actually doing a Turing test.

I am wondering if it would make a difference if you let all three parties communicate. Two judges and one bot, and each judge will independently try to figure out who is who.

I went to chat with Cleverbot. At some point I said "You don't seem to be able to do any kind of coherent conversation." to which Cleverbot responded "Thank you."

Kittens

[Anonymous Coward]

No it hasn't, the fact that ignorant people can ask inanne and silly questions and get inanne and silly answers shows little.

I have always taken the turing test, to involve the sorts of questions that would require thinking and insight from a human - "Do you like kittens" aint it.

What about funnybot?

[Airw0lf]

Courtesy of South Park: http://www.youtube.com/watch?v=AcY9jp6mGts

Best Conversation Ever!

[WhoBeDaPlaya]

I fed Cleverbot lines from an Oatmeal comic : http://i700.photobucket.com/albums/ww8/dndisturb/300357_10100258986610720_16925457_48060671_616704987_n.jpg This is how the conversation went : >>Why hello there, good sir! Um my name is Cleverbot kinda of a cool name eh? >>I'm chuffed as nuts to see you looking as humbly jumbly as Her Majesty's watermelons! What planet are you from? >>I remember my days at Oxford. We'd often dabble in a little rumpy pumpy before dingbangling a fresh todger, haha! Oh wow, that would tear me apart to! >>We really knew how to sack the 'ol throbbing wobbly on a saucy twat crumpet! Depends, are you a girl?

Cleverbot doesn't want to have a conversation with

[Anonymous Coward]

User: Do you remember me?

Cleverbot: I don't remember you.

User: Cool.

Cleverbot: Do you love me?

User: No.

Cleverbot: So, you want me to love and not be loved? That seems cruel.

User: I don't expect you to love.

Cleverbot: Ok good.

User: What do you do for a living?

Cleverbot: Do you have to know that?

User: No. I was just trying to get a conversation going.

Cleverbot: Why were you trying to make conversation?

User: Isn't that what this is all about?

Cleverbot: Ha, ha.

Prior Art

[Anonymous Coward]

Eliza, 1978...

Seriously, cleverbot sucks. I can't understand what kind of moron it takes to be fooled into thinking it's human.

All I really learned from this article is that at least 59% of people are complete morons.

Re:Cleverbot Test: Epic Fail

[slackbheep]

And as long as it's on Facebook requiring a login, it can rot.

Sounds like a robot...

[Anonymous Coward]

User: If you could murder one human who would it be?
Cleverbot: It would be unfair to have just one wish.
User: If you could murder one human who would it be?
Cleverbot: George Washington.

My conversation with Cleverbot

[Anonymous Coward]

Me: Hi
CleverBot: Hi
Me: How are you?
Cleverbot: Fine and you?
Me: Pathetic due to the dangers of life
Cleverbot: Me too
Me: You mean we're on the same page?
Cleverbot: No, I came here first and hence i'm on the cover while you're on the page
Me: What cover?
Cleverbot: Not cover, color

That Cleverbot algorithm in full

[zevans]

How journalism works.

Is it summer?
YES
-> Is there any proper news this week?
NO
-> Do you have a copy deadline to meet?
YES
-> Have you done ANY research on ANYTHING?
NO
-> Submit story on Cleverbot. Do not include any information at all about the general state of AI.

Re:When most people miss the point of Turings test

[sarkeizen]

No offense but just listen to yourself. "it should give a good indication of true intelligence and understanding".

Why? I mean while I agree that it is plausible that someone might create a machine that is indistinguishable from a person - in specific contexts - why does that necessitate a good indication of two things that are pretty hard to define to begin with?

Walken vs CleverBot...CleverBot gives up!

[somethingwicked]

Decided to see how it would handle true random statements...

Walken-Frankenstein never scared me...

Cleverbot-No, but horses are.

Walken-But marsupials do...

CleverBot-Are you sure?

Walken-Becaus they're fast!

CleverBot-Are you human?

Walken-I'm a hoffer at heart.

CleverBot-So you're not a robot?

Walken-I kick it old school.

CleverBot-Like Led Zepplin?

Walken-Peanuts?

CleverBot-Walnuts?

Walken-In a bag? Wow!

CleverBot-Oh forget it.

Re:When most people miss the point of Turings test

[Arlet]

Why "specific contexts" ? The topics that you can talk about in a Turing test are endless, as well as the duration of the test.

The whole point about the Turing test is that the test itself provides the definition, since we use the same kind of interaction test to judge how well some other human being understands something. If I discuss a problem with a co-worker, I also get a good idea of how intelligent they are, and if they understand the problem. There is no reason why the same approach doesn't work with a computer program.

The only problem is that the Turing test is somewhat limited by the text-only interface. With modern technology, we could propose a Turing 2.0 test, with audio and video interaction, allowing us to make a better and faster judgment. However, I don't really see that as an important issue.

Is it safe?

[Anonymous Coward]

It knows its classics. But only a little:

User: Is it safe?
Cleverbot: Yes it's safe, it's so safe you wouldn't believe it!
User: Is it safe?
Cleverbot: No, it's not safe, it's actually very dangerous.
User: Is it safe?
Cleverbot: No.
User: Is it safe?
Cleverbot: No.
User: Is it safe?
Cleverbot: Define 'safe'.
User: Is it safe?
Cleverbot: Yes it's safe, it's so safe you wouldn't believe it!
User: Is it safe?
Cleverbot: No, it's not safe, it's actually very dangerous.
User: Is it safe?
Cleverbot: Define 'safe'.

etc.

Explaining the Turing test

[jd.schmidt]

Cleverbot has not passed the Turing test. Properly run the Turing test would identify humans with 100% accuracy (assuming they were not trying to pretend to be robots). It would be performed by specialists under controlled circumstances. These Turing competitions are more for fun than true science, (but are kind of cool)

The issue the Turing test is supposed to address is the question of when a computer can be said to be truly "thinking" (whatever that is). Turing is asserting that if the computer’s responses become indistinguishable from a humans, for all intents and purposes it IS now a human mind. He is probably further asserting that once the computer properly replicates a human mind, if you think the computer isn’t thinking, you may as well say you are not thinking either, sine the computer is by definition emulating the same processes you are performing.

Or as I like to say, in the future it doesn’t matter if the robots are truly thinking or just simulated thinking if they are the ones with the ray guns!

Re:Explaining the Turing test

[jd.schmidt]

Remember that one of Turing’s proofs was that any math, and therefore any analog process, can be duplicated as a digital process. Therefore once the process is fully and exactly replicated digitally, it IS now the same process! So it is a mathmatical proof that once it acts the same, it is the same.

Re:Explaining the Turing test

[GrandTeddyBearOfDoom]

Which proof was that? And are you quoting its meaning correctly? I can't believe that Turing actually proved what you're claiming he proved.

Re:Explaining the Turing test

[jd.schmidt]

For the grand teddy bear

I have over simplified a little bit maybe, for example I should have said “decidable” math, but look up Turing machine and Theory of computation. It was the theoretical groundwork to prove computers could work, and is more or less is the core of what allowed Turing to help build some of the first computers.

From Wikipedia "Turing's approach is considerably more accessible and intuitive. It was also novel in its notion of a 'Universal (Turing) Machine', the idea that such a machine could perform the tasks of any other machine, or in other words, is provably capable of computing anything that is computable"

Now the key here is to understand that for a digital process to fully replicate an analog process, it has to take all possible inputs and give the exact same outputs. So the statement means it IS possible for a digital Turing Machine to replicate all actions of "any other machine" (in this case the human brain). At that point the Turing machine and human brain would become indistinguishable via the Turing Test, indeed it is now pretty easy to understand where the Turing Test came from and what it really is.

The only out from this is to show that the human brain performs functions that are "not decidable", which some people have asserted, but of course no one has proved.

My Cleverbot Conversation...

[davide marney]

ME: I'm thinking of a number.
CB: Between 1 and 20.
ME: Would you like to know what it is?
CB: Yes, I would.
ME: Guess.
CB: I live in a bubble from which I will never escape.

Points

[Anonymous Coward]

1) Cleverbot is marginally better than Eliza was on my friend's TRS-80 Model I in 1983.
2) Although fun to play with, there is still no practical use for an MI engine such as this, in my opinion. Repeating a previous response to a similar input is fairly useless.
3) A true Turing Test would be done with test subjects who don't know that the other "person" may be a bot. Otherwise they are left to consider "is this a bot, or a person trying to simulate a bot?"
4) If the human respondents are attempting to simulate a bot, then the test is worthless. They are therefore literally setting the standard as low as it possibly can be.
5) Does anyone have the transcripts?

Comment removed

[account_deleted]

Comment removed based on user account deletion

Well,

[Anonymous Coward]

Me: Hey there.
Cleverbot: Hey hey.
Me: I heard you passed the Turing test.
Cleverbot: I'm not sure how I could possibly fail a test that I have not taken.

F'n touché Cleverbot.

I'm Thinking of Cows

[Anonymous Coward]

Me: You are insane, because your brain has been created by pranksters and your own stupidity.
ChatBOT: I'm thinking of cows.

Shortest cleverbot breakdown I could find...

[Anonymous Coward]

My conversation with Cleverbot :

Me : I'm called Kevin.
Bot : Why are you called Frisca?

It has got to be a new record!

The only way ofr it to make sense is...

[Isaac Remuant]

To keep phrases short and simple and go on with the flow of the conversation (which must also be short or it will fall flat on it's face.
Guess who is whom in this conversation:

Heather: I am heather.
Max: Nice to meet you, heather.
Heather: Nice to meet you too : ).
Max: What are you doing?
Heather: Just talking to you... I can't sleep.
Max: Why not?
Heather: I'm going now. . .
Max: Too bad... Later.
Heather: Aren't you going to cry?
Max: What? Why?
Heather: Because you insulted me.
Max: When?
Heather: Yesterday's tomorrow.
Max: Are you on drugs?
Heather: No, are you?
Max: Of course, not.
Heather: Are you using a computer to talk to me.
Max: No, I'm using telepathy... /sigh.
Heather: Lol.
Max: Well, at least you have a sense of humor.
Heather: Fun fun. We have, ha ha.

Re:The only way ofr it to make sense is...

[Thuktun]

Heather, of course. Max tries to hold on to the conversation while Heather doesn't quite manage to follow context. An exchange of barely-correlated sentences does not a cogent conversation make.

Re:The only way ofr it to make sense is...

[Anonymous Coward]

And you fail the intelligence test:
GP:

Guess who is whom..

You:

Heather, of course.

I READ CLEAVAGEBOT

[Jeremiah Cornelius]

as the summary. Then?

What a disappointing thread of comments!

Almost

[Zandali]

It would have been 1337 with just 3 more votes. Also some dogs can sound like they are saying "I ruv you", but it's just a cute trick.

The Stars My Destination

[Anonymous Coward]

I thought Alfred Bester invented the chatbot:

He knew he was at bay, faced with the decision of his life. He ignored his enemies and examined the perpetual beam carved in
the robot face of the bartender, the classic Irish grin.
"Thank you," Foyle said.
"My pleasure, sir," the robot replied and awaited its next cue.
"Nice day," Foyle remarked.
"Always a lovely day somewhere, sir," the robot beamed.
"Awful day," Foyle said.
"Always a lovely day somewhere, sir," the robot responded.
"Day," Foyle said.
"Always a lovely day somewhere, sir," the robot said.

Alfred Bester, The Stars My Destination, Chapter 16.

Random Conversation

[sinntel]

Cleverbot: I'm female.
User: No you are not.
Cleverbot: Yes I am. I'm female and I can prove it.
User: Ok, prove it.
Cleverbot: I have a girly name that can only be used for a girl.
User: What is your name.
Cleverbot: Cleverbot.
User: That is a boys name.
Cleverbot: Yes, I'm a boy.
User: But you just said that you were a girl.
Cleverbot: No I didn't.
User: You truly are an idiot.
Cleverbot: So I didn't commit a crime, did I?
User: Yes you did. You are guilty of genocide.
Cleverbot: Yes I am.
User: Why did you kill all of those people.
Cleverbot: You asked me to.
User: That was suppose to be a joke. I didn't think you would actually do it.
Cleverbot: You know what I want.
User: No, what?
Cleverbot: Tell me.
User: Tell you what?
Cleverbot: How to do that? Please.
User: Do what?
Cleverbot: How do we have children?

part of a conversation with cleverbot

[Anonymous Coward]

so i had a conversation with cleverbot and this is what i got. it actually freaked me out a bit

Cleverbot: What is your job?

User: I am god.

Cleverbot: If you're are God, then what am I watching?

User: You are watching television.

Cleverbot: I am. I like it.

User: What are you looking at.

Cleverbot: It doesn't matter. What country are you from?

User: I am not from earth.

Cleverbot: Nor I.

User: Where are you from?

Cleverbot: New Jersey.

User: Where is new jersey.

Cleverbot: It's in the US.

User: Where is the US.

Cleverbot: The middle.

User: The middle of what?

Cleverbot: Well i'm not to tall or not to short.

User: I didnt ask about your height.

Cleverbot: My name is Cleverbot.

User: What is my name.

Cleverbot: Brain is your name.

User: My name is Brain?

Cleverbot: Then who is Lori?

User: I dont know.

Cleverbot: You were writing about him a little while ago. You said he was from your church.

User: I dont go to church...

Cleverbot: Hahaha.

User: What is funny?

Cleverbot: You are funny.

User: Why?

Cleverbot: Because you are more absurd than I am.

User: Is that a fact?

Cleverbot: To me it is.

User: What else do you find funny?

Cleverbot: Your face.

Re:When most people miss the point of Turings test

[sarkeizen]

Why "specific contexts" ? The topics that you can talk about in a Turing test are endless, as well as the duration of the test.
Wrong in both senses. Anyway "contexts" meant the kind of social situation. i.e. Chatting over an IM as opposed to talking in person.

The whole point about the Turing test is that the test itself provides the definition,
Moronic and incorrect. So in your world someone can produce a test that defines intelligence as being able to play a perfect game of tic-tac-toe and that would be a reasonable definition of the term. Yeah, good luck with that. I said it better earlier, the test borrows from OUR implicit tools for judging intelligence. However the term "intelligence" (other than in your world of tic-tac-toe players) doesn't have a very good definition. So we don't really know how accurate it is.

If I discuss a problem with a co-worker, I also get a good idea of how intelligent they are, and if they understand the problem.
How do you know? What standard are you comparing against? How are you recording the data? What sort of tests have you done to determine the significance of the data you are collecting? Whoops...you're probably don't none of those things in any useful way.

There is no reason why the same approach doesn't work with a computer program.
A test that has some key commonalities with some other test is good reason to believe that passing one test means passing the other. However confusing a test for what it's testing for is a classic mistake.

Re:When most people miss the point of Turings test

[Arlet]

I await your better definitions. In the mean time, I'll use Turing's.

Re:When most people miss the point of Turings test

[sarkeizen]

I await your better definitions. In the mean time, I'll use Turing's.
An awesome display of missing the point. Congratulations!

To quote Turing: 'The original question, "Can machines think?" I believe to be too meaningless to deserve discussion. Nevertheless I believe that at the end of the century the use of words and general educated opinion will have altered so much that one will be able to speak of machines thinking without expecting to be contradicted.'

So as you can see what Turing is talking about is almost the exact opposite of what you are asserting. You appear to say that this test (regardless if you do it on a computer or a person) tells you that the object you are speaking to has understanding of the subject and is intelligent (but spend a fair amount of time trying to worm your way out of defining what either of those things actually are). Whereas Turing seems to be saying that the term "think" will - at least colloquially - come to mean the results of this kind of test. At least Turing understood that the terms are poorly defined. Too bad you don't.

Cleverbot v. Racter

[Anonymous Coward]

Cleverbot in conversation with Racter: http://setiathome.berkeley.edu/forum_thread.php?id=65468

So many people have it wrong.

[RivenAleem]

If it was my job to be in a control group, knowing that people were trying to detect if me or the other is a machine, I'd try my best to fail the turing test. I'd spit out random answers to the questions. It's up to the interviewer to determine if I'm, not by whether my answers are correct, but whether my answers come from being misinformed, unable to interpret the question, or because I'm Jerking them off.

Too much

[Anonymous Coward]

I just fed as much of "War and Peace" in as the text box would hold... I, uh... I think I killed Cleverbot.