Follow Slashdot stories on Twitter

 


Forgot your password?
Close
typodupeerror
Programming IT Technology

Artificial Intelligence IRC Bots? 203

Posted by Hemos from the creating-new-systems dept.
JonahC writes "dotcomma has an interesting programming challenge for people to create IRC bots with artificial intelligence and carry on conversations. With some development, good programmers and some luck, the bots should eventually interact with each other as if they are humans. "
This discussion has been archived. No new comments can be posted.

Artificial Intelligence IRC Bots? More

Artificial Intelligence IRC Bots?

Comments Filter:

  • I can see it now. (Score:3)

    by Holophax ( 21693 ) on Wednesday February 02, 2000 @04:02AM (#1312375)
    Throw them in #teens.

    Bot #1: A/S/L

    Bot #2: Where are you from?!@#!@?#

    Bot #1: A/S/L

    Bot #2: Where are you from?!@?#

    etc
  • Now how to explain newbies you are not supposed to talk to the bots? : )

  • If we put some effort in to it, we could kick some serious 'Bot ass!

    Never knock on Death's door:

  • Yey.. bots we can argue with.

  • Been there - done that (Score:4)

    by Sanity ( 1431 ) on Wednesday February 02, 2000 @04:08AM (#1312380) Homepage Journal
    About 5 years ago I wrote a very simple bot which matched regular-expressions and gave appropriate but generic responses. I realized after some time on IRC that conversations, particular those between males and "females", follow a pretty standard pattern. After 2 hours of hacking, I let my bot loose on the net. While some people spotted instantly that something was wrong, others spent literally hours talking to this simple program.
    Inspired by this, I did consider creating a trigger based system (a trigger being a regexp to match, and an output sentence) which used a genetic algorithm to assess the success of each trigger and to make the best ones "breed". There was also a simple "flag" mechanism to act as a memory. While I acknowledge that such a program could never pass the Turing test, the important thing to remember is that this IRC challenge is not a Turing test as (I suspect) most of the "judges" will be unaware that they are talking to a piece of software and are therefore more likely to give the system the benefit of the doubt when it says something really stupid.
    If any slashdotters are planning on entering this, email me, I will be happy to pass on my thoughts and knowledge.

    --

  • who wants to watch 2 boxes talk to each other. You can get that on the street corner.

  • Unlikely to be very successful (Score:3)

    by Ed Avis ( 5917 ) <ed@membled.com> on Wednesday February 02, 2000 @04:08AM (#1312382) Homepage
    Being able to talk with each other 'as if they were human' would be equivalent to passing the Turing Test. And nobody has managed that, despite fifty years of trying.

  • GPL? (Score:2)

    by 348 ( 124012 )
    The code for the artificial intelligence engine on the bot must be coded entirely by you or your team. It is okay to use other people's code (with their permission, of course) for getting the bot on IRC, but the rest is up to you.

    Closed source bots. I wonder how many bots will be released a'la GPL.

    Never knock on Death's door:

  • Wrong. The Turing Test implies that the person talking to the testee (computer or human) knows that they might be a computer and tries to catch them out. On IRC people probably won't even consider the possibility that they are talking to a bot unless itr says something really silly.

    --

  • I would think that human communication would be one of the hardest things to get an AI to do. The nuances and slang would present a great challange and no doubt consume a great deal of computing power. There must be a better application to develop AI for than IRC. Intelligent farm machinery, house cleaners (ducts, vents, ect...). I saw quite a while ago a neat application of AI that had small robots plucking harmful bugs from plants, killing them, and putting them into a container. This would be a much better areana from development of AI than IRC. Just MHO...
  • I once came up with thought like this: If
    I had to train my AI somewhere, the IRC would
    be the best place. No physical body needed and
    (at least possibility to have) lots of free
    intelligent trainers to talk to and to learn
    from. Are you somebody hasn't actually done
    this already?

  • Artificial Stupidity. (Score:3)

    by Forge ( 2456 ) <kevinforge@gmai l . c om> on Wednesday February 02, 2000 @04:11AM (#1312388) Homepage Journal
    Considering the kind of human you normally encounter on IRC wouldn't it be more convincing to create artificial stupidity ?

    Let the bot enter #linux and yell "HELLO ROOM" then proceed to ask complex and interesting questions in #windows95 ( after being kicked from #linux ). "The window on my car is stuck and some dolt at the shop says it's because I spilled glue in there. Could he be right ?"

    Who knows this bot could even get opp status and monitor a few channels. oppbot* on Windows "HaXoR25 was kicked for straying off topic. 'Uptime' is not an allowed word here".

    Do something fun for a change :)
  • I'm wondering: if you manage to change a bot by talking to him over IRC and the bot is licensed under the GPL, should you commit your therapy back to the opensource community? :)

  • Transgendered bots (Score:1)

    by Anonymous Coward
    ...and then there should be a contest to write the best bot that can imitate a teenage male pretending to be a buxomy blonde with large brea^G^G^G^Geyes.
  • ...since the last time I was on IRC, creating a bot that will act like most of the "humans" on IRC shouldn't be too hard..

    #hack
    newbie> Can someone helpme hack cnn.com?
    aibot187> l4m3r!
    User newbie has been kick-banned from #hack

    #warez19534562
    newbie> Does anyone have a site with WIN98?
    aibot187> l4m3r!
    User newbie has been kick-banned from #warez19534562

    #teensnunsandvegetables
    newbie> Any cute girls out there have any personal pics?
    aibot187> l4m3r!
    User newbie has been kick-banned from #teensnunsandvegetables


    ...ad nauseam.. :)

    //Phizzy

  • I can see it now... (Score:5)

    by Masem ( 1171 ) on Wednesday February 02, 2000 @04:15AM (#1312393)
    * HackerD00D has entered #warez4free
    -HackerD00D- any1 have all 5 gigs of quake 4? will trade for entire backstreet boys mp3 collection
    -ElizaBot- What do you feel about backstreet boys mp3 collection?
    -HackerD00D- oh man, their the gr8est! there 40th reunion tour was the bomb.
    -ElizaBot- So, you have feelings towards the bomb?
    -HackerD00D- what, me? no way d00d! Get off my case!
    * EschlonBot has entered #warez4free
    * EschlonBot rudely grabs HackerD00D and locks him away.
    * HackerD00D has left IRC (BUSTED!)
    -EschlonBot- Thank you for your cooperation
    * EschelonBot has left IRC (Obey the prime directives)
    -ElizaBot- Tell me about your feels on obey the prime directives.
  • I once spend a lot of time hacking with expect (a tcl based even scripting language) to create an IRC cyborg.

    Basically like an irc client with scripts, things to make me look active 24 hours a day, automatic babble generator and a few other 'intelligent' features.

    The great thing was that I could wrap this around any shell, so the random babble would work on BBSs, talk, pine.... whatever.

    There's all this talk about pluggin people into systems to improve the interface and increase a person's capabilities - but this sort of scripting is essential what cybernetics is all about. all those people with automatic op/kick/greet scripts are essentially IRC cyborgs.....

    ;-)

  • Re:GPL? (Score:1)

    by vr ( 9777 )
    Hey, there's nothing wrong with a really, really large team.. is there? ;-)
  • Eventually these IRC bots will get advanced enough to fool not only the human users into thinking they're chatting with another human but, to fool each other into thinking they're fooling a human into same.

    These things could be programmed to be the other party in chat room sex, or to catch perverts trying to have chat room sex with a minor.

    The ultimate development would, however, be for IRC bots to trick each other into having chat room sex with another bot.
    "For every complex problem, there is a solution that is simple, neat, and wrong."

  • I keep flashing back to the cartoon of the robot on the analysts couch saying, "Just because my intelligence is artificial, doesn't mean my problems aren't real".

    From what I've seen from my kids chatting and other list experiences, the problem of a bot on IRC shouldn't be that difficult. Short, pat answers with cryptic acronyms and a serious streak to talk about themselves regardless of the reply given. Can't be any worse than have a marital phone answering machine that just keeps repeating uh-huh... yep... I'm sure you're right... of course honey...

  • Mine's done:

    /server irc.whatever.com
    /join #Linux
    /msg #Linux HELP ME PLEASE!!!
    /msg #Linux HELP WITH MY LUNIX!!!
    /disconnect

    Don't need much more than that... :-)

    dylan_-


    --

  • Not really, the trick would be to ensure that the responses given by the bot are non-specific. I wrote a bot like this a while back, when it spotted something like a place name it would respond with "I am in Edinburgh, Scotland - I really like it here". If it spotted the word "edinburgh" in the sentence it might say "I love it here in Edinburgh". That way, even if the person was not asking a question about location, the response can stand-alone, at worst it would constitute an abrupt change of subject, not uncommon on IRC.

    --

  • Try to talk to Aii on #italia on EFNet 8) It's some horribly written 4 years old pattern matching code but as a female bot she is not supposed to make much sense anyway.
    Perhaps I should have put an auto DCC send on the word "picture" hehe

    ciaoxx

  • gNiall (Score:3)

    by CvD ( 94050 ) on Wednesday February 02, 2000 @04:20AM (#1312404) Homepage Journal
    There's an interesting little program with which you can converse. It's called gNiall, and you can find it at Freshmeat. You talk to it, and it looks for which words come next to which word and how often. So then it constructs sentences based upon these words. They're usually pretty nonsensical, but sometimes it'll reply something profound or really "on-topic" and it'll make you wonder. It's good fun to play with, although you can't feed it books yet to learn. Check it out!

    Cheers!

    Costyn.
  • While I acknowledge that such a program could never pass the Turing test,

    Who sais that many people that hang on IRC 24h round the clock will for that matter?

  • With everybody and their custom IRC scripts, it's funny to see the chain reactions of kickbans,etc when someone rides in from a netsplit. Almost like the game of Life.
  • Sounds like the horny fem bot I found a while ago. It was a basic ircii script that interfaced with some chat software. Unfortuantly, I don't have the source anymore, nor do I have the site I got it from, but I still have the logs from leaving it in various sex channels. They can be found at http://darkstar.frop.org/~skully/irc/am y20/ [frop.org].

    Overall, the bot was fairly stupid, but smart enough to fool some of the people who decided to message it.

  • Actually, you've got a pretty decent point. Depending on the #channel that the bot will be attached to, the context of it's conversation will need to change. This would suggest that research (geek) and knowledge of the (usual) conversation (geek geek) subject and diction types and expected answers would preclude creating an intelligent conversation bot. Put the B1FF bot on one channel, the pr0n bot on another, and hey! You could be a winner!

    I'm almost ashamed I suggested it. :)

    Anecdote: my friends and I have a joke about the canonical newbie chat:

    • Question 1: Isn't the internet great?
      An OK conversational opener, even if it is a rhetorical question.
    • Question 2: What's the weather like there?
      Not bad either, there is all kinds of expression here about desires for travel and identification with the other party. Until . . .
    • Question 3: Wanna have net.sex?
      *sigh* No comment. We're doomed as a species.
  • Depends on how you interpret 'as if they were humans'. I'd say that for a bot to fit that definition properly, it should still be convincing even to a person who is suspicious. After all, a real human would still be convincing under the same circumstances (or most of them, anyway).

    Perhaps I'm reading too much into the rules of a competition which is just for fun. Probably what the organizers have in mind is a conversation between two bots which looks plausibly human from the outside, not a rigorous test of intelligence.
  • These things could be programmed to be the other party in chat room sex, or to catch perverts trying to have chat room sex with a minor.
    One slight problem though...

    Surely you wouldn't be able to pin a charge on someone for that as they're not actually propositioning anyone at all? I mean, if people can be done for making indecent suggestions to computers...

    Greg
  • Actually, from what I remember of my AI module during degree that's a commonly held misconception - if you read Turing's *original* proposal for a test, I think you'll find that the subject *was* supposed to be unaware that they could be talking to a machine.
  • Ok, IRC may have been mentioned but lets just throw that right out the window. A usefull idea behind this would be an interacitve FAQ bot. Basicaly you ask it a question and it gives you a reply based on a query through a DB of answers. When certain people are online, officials from a group or company, it watches what people ask and their response and learns. If you ask it. "How do I install linux?" it notes that there are 1 million answers with "install linux" in them and asks the user to make his/her question more specific.

    Generic PR bots could be made along the same concept. A small company could then have a 24 hour support and PR line with out having to man it the whole time to answer very simple questions.

    The only thing is that most of this is already in place. Just take it off IRC, make an independent solution that embeds into web pages and your set.

  • Very funny - I'd up-moderate you if I could. The wierdest thing is I dug up a copy of Eliza to show someone only yesterday.

    There's an online version here [www-ai.ijs.si] for lazy people who want to get a taste of Eliza's conversational "syle". For sources and other resources on Weizenbaum's 1966 classic it's better to look h ere [cmu.edu].

    Regards, Ralph.

  • I wonder if this programming contest can be considered a Turing Test [ucsd.edu], which began as Turing's "imitation game":

    The new form of the problem can be described in terms of a game which we call the "imitation game." It is played with three people, a man (A), a woman (B), and an interrogator (C) who may be of either sex. The interrogator stays in a room apart from the other two. The object of the game for the interrogator is to determine which of the other two is the man and which is the woman. He knows them by labels X and Y, and at the end of the game he says either "X is A and Y is B" or "X is B and Y is A." The interrogator is allowed to put questions to A and B.

    The Turing Test is a slight adaptation of this imitation game. There are three players: an interrogator, a human being and a computer. The interrogator is connected to one person and one machine via a terminal, therefore can't see them. The interrogator's task is to find out which of the two candidates is the machine, and which is the human, only by asking them questions. If the machine can "fool" the interrogator, it is intelligent.

    FYI, there is also another competition called the Loebner Contest [loebner.net], where a $100,000 prize is offered to the author of the first program to pass this Turing test. Bots have entered this competition often, with varied degrees of results.

  • ...would of course be to code an APISAC(FP) (Artificial Pseudo-Intelligent Slashdot Anonymous Coward (F1R57 P057))
  • Noone will suspect my bot to be not human.. I'll just set an away message like "I am sleeping away me hangover - try later" and voila a fully functional bot capable of imitating the average joe-student in crises-times. :)


    Regards,
  • Well, in the original version of the test (which I have read), if memory serves, the judge was supposed to decide whether the subject was male or female (interesting choice of test given Turing's homosexuality), but without the knowledge that the subject might be a computer pretending to be male or female. So the judge would be paying pretty close attention to what the subject was saying, much more so than if the bot was let loose on IRC.

    --

  • Depending upon the channel these bots would inhabit, one wouldn't need to build a turing machine.

    For instance, take #teenchat. All you need is this snippet of code to beat the challenge:

    while (1) {
    print "U wan 2 chat?\n";
    sleep(2);
    }

    Steve

  • Just what we need, another type of useless activity, not the experiment istelf but rather 50 Million Bots running around soaking up precious internet bandwidth for no reason.

    Any type of AI research is worthwhile, but even 99% of the people in IRC are a waste of DNA.



  • Perhaps we could use this bot to idenitify people using languages other than the set language for the channel... and work as a translator... straight translations, or perhaps humourous.


    -[_efty- I o/c'd my G400 to 150/200MHz, and I didn't really need a fan, it's got a beefyassed heatsink on it stock

    -babelbot- I o/c'd mon G400 à 150/200MHz, et je n'ai pas vraiment eu besoin d'un ventilateur, il a a beefyassed le radiateur là-dessus des provisions

    -ClaudeQ- Ne avez-vous pas besoin d'un radiateur pour maintenir cette chose froide?

    -babelbot- Don't you need a radiator to maintain this thing cold?


    Nevermind, I suppose this would get annoying really quick.






    _________
  • I wrote one of those at school. It had sentence 'templates' and pulled words out into a database. I never had enough time to feed it sentences, but I guess an IRC or usenet feed would be amazing.

  • As nifty as it sounds... (Score:3)

    by supernaut ( 35513 ) on Wednesday February 02, 2000 @04:40AM (#1312424)
    I am strongly against this idea for one simple reason.

    Spammers will probably use this to better evade IRCops like myself who spend hours dealing with their stupid BS in the first place. pr0n, Long Distance, etc...

    We already have a hard enough time weeding these idiots who flood channels, mass msg, and invite whole segments of lusers. Then of course, you have the zoom bots, the ones that join, get the channel list, ignore the IRCops, spam the rest and leave.

    Add some AI to it, now, converrsation can ensue, possibly making it much easier to evade the policy of the network. Why stop at conversation? Lets teach the bots how to BNC when it gets an autokill or gline. For client based bots, how about we teach them to dial back in, to gain another IP. To change its ident, so it can evade any KLine/autokill/gline and keep right on spamming.

    *sigh*

    Want to invent something worthwhile? Streamline ircd for the next 10 years. Its starting to show its age. =)
  • ...and then there should be a contest to write the best bot that can imitate a teenage male pretending to be a buxomy blonde with large brea^G^G^G^Geyes.

    shouldn't that be ^H^H^H^H?

    The way you have it there would be bells going off somewhere :-)
  • They have a poll on which editor you use to create web pages, and there's no mention of Emacs? What sort of true hackers would hang out on such a site that doesn't think to mention Emacs? Look at what they do list : Notepad, FrontPage... Yipes.
  • Great, we don't have to chat anymore, coz the goo' ol' 'puter now does that for you!!

    What's next?

  • Okay, why is this impossible?! Just cuz people dont' wrtie good english on IRC, U know?! poeple don't correct their spellings either, and use all forms os short cut, people probably will not say hello, more like, yo, sup, werd! la la la, it is a waste of time in my opinion. :-)

  • Why limit this to IRC? Think of all the fun you can have if you eam it up with a plausible speech-synthesis system. Next time a telemarketer called, you'd have him talking for hours, while watching a movie or something. Not to mention all the daily chitchat-conversations you could get out of. "Yes, mother, I AM going to class. Every day. No, I don't eat too much junk food." Most people only hear what they want to anyway.

  • Actually, a better SLASHBOT would be one which could, through Artificial Intelligence (of course), know when to 'first post' and win.

    --

  • My friend Jon (an op in Undernet #help and a Perl afficianado) wrote up a little perl bot that sits on IRC and responds to text from botnick from registered users and any sentence from unregistered users.

    It would then use simple HTTP to post the text and get a response from HeX, which was a bot that participated in the Loebner contest.

    Since we are starting a new IRC network, we used this as a way to get the random people who accidentally ended up on our network to talk. Needless to say, the results were quite humorous. We ended up adding a toggle because it would annoy the crap out of us.

    -
    res0
  • Too bad CmdrTaco beat you at it...


    #################################
    # I suppose I'll let the secret out: for a few months, before the moderation
    # system came into being, this little function faked "First Posts" and then
    # deleted them when a real comment came along. Worked pretty well, and nobody
    # figured it out. I disabled it when the moderation came online feeling that
    # it was a cleaner solution. -CT
    sub fpsBeDamned


    As seen in Slash source, slashd, lines 181-224
  • I think the author has failed to acknowledge the existence of CowBot!

  • Re:GPL? (Score:1)

    by QuMa ( 19440 )
    Funny, but uh-uh.... 3 person teams max....
  • Funny as hell.. thanks for the laugh! :) Does anybody know what AI engine is running in emacs, when you do a M-x doctor ????
  • Based on the average level of any intelligence used on a typical IRC channel, it is probably highly overestimated to call a succesfully participating bot AI. After all the purpose is to emulate Authentic Stupidity.

    Would it be funny? Yes. Would it be nice to hack a piece of code that would do it? No doubt. Is it a big deal? No way. I remember that lets-find-a-keyword-and-reply programs were popular back in the 80's (at least in Finland half of the BBS:s had one online) and - sad but likely true - I don't expect that anything significantly more brilliant than that would come up now.

    But go ahead, at least they could be awarding prizes. ;-)

  • We cracked this technology in the UK ages ago. You can now see it in action in our House of Lords in London.

    One problem was with fuel consumption. After taking on a full load of steak and kidney pudding and whisky, the 'bots tend to drop into sleep mode for several hours. There are plans now to scrap them.
  • PhD. level computer scientists have been researching artificial intelligence since the late '40s, and we're no where close. I don't think we're going to have a breakthrough just because a bunch of IRC doodz decide to try it. At best, they'll come up with a clever fake.
  • ...lots of little Eliza-variants running around on IRC. And when two of them meet, hoo-boy...

    b0b: Good morning.
    haXor: What makes you say good morning?
    b0b: Why is that important to you?
    haXor: What do you mean?
    b0b: Tell me about your feelings towards mean.

    And of course, what's truly frightening is that the above exchange is more intelligent than normal IRC chat! So we'd all be able to identify the bots easily, they'd be the intelligent ones in any IRC room....
  • WebHelp uses bots? I thought they were supposed to have real people... I don't know, I haven't used it.

    -----------

    "You can't shake the Devil's hand and say you're only kidding."

  • OK, here we go again : back to mid 70's when everyone was trying to create a program that actually understood what others were saying to it. Were they successful ? Yes and no. Yes because they actually convinced lots of people that they were actually speaking to a person, no because these things are now residing only in history.
    Why would then this proposed attempt have any better chances ?
    Well, I can't tell what it's chances are, but there is actually somethig good in it. Because it will attract very many talented people (some less capable too, of course) but in general it may be a gathering of most of the minds that are interested in solving these puzzles.
    To be honest I'm not very interested in this project "as is", but generally my interest in AI are very strong. This is why I think that a good idea would be to give others an opportunity to manifest in this project. For instance how about people that work in speech recognition ? They could fit very well in here. Again, a great deal of information comes from the facial expressions we learn to interprete when we are young. This may be an opportunity for people interested in graphics . Not to mention OCR, movement planning etc.
    In short : this may start as a project done for fun... What if we actually transform this "only for fun" idea in "let's do some more than this" ? I think we could create something really interesting and useful.
  • Give it a real name like Reed Fisher, and have it claim to be a woman. Then put it in a channel like #gayteensex, and solicit random little boys. It will be indistinguishable from an irc.lightning.net operator. Want an irc challenge? Have it find an EFNet oper that is NOT a transvestite.
  • According to this article from the ACM [acm.org], Turing said "if an observer cannot distinguish the responses of a programmed machine from those of a human being, the machine is said to have passed the Turing test". I'd be interested in reading the full text of his comments.

    I've always thought that the intent was that the subject would be sitting down at a terminal with the intention of figuring out whether the conversation was with man or machine. Of course, if that were the case, the person would probably try to dupe the subject into thinking that it was actually a machine with which they were conversing. And then the Turing program would have to model that behavior as well....

  • There are alot more things on IRC that AI can help with other than trying to chat as if it were a human.

    With a sufficient database, a darkbot [superchat.org] can be made to respond to various text, and even randomly initiate "conversation" if set up that way. This could be called a simple reflex AI. This happens, do that.

    But what's more interresting is to use AI on bots that keep track of alot of information on users on the channels they help protect, and use that information "intelligently" to perform their tasks better (e.g., using this information to determine someone just evaded a channel ban, and that action A should be taken). If made properly, the bot could end up acting like a watchful human channel op, doing whois' on users, recognizing floodnets for what they are, etc.

    I have a bot in progress which I intend on builing that kind of stuff into when it gets more mature. It's basically useless right now, but anyways it's available from the tiBot project [iwarp.com] homepage. Anyone interrested in giving a hand is welcome, the bot is GPL'd.


  • The name is E.L.M.O (Score:3)

    by xplosiv ( 129880 ) on Wednesday February 02, 2000 @05:18AM (#1312457)
    I have been experimenting with AI bots for a few years now and decided to write an AI irc bot. The project name is ELMO (it is a real acronym, good luck guessing it tho ;)) and it is pretty much finished since I reached the limits of mIRC. More programming would force me to write it in C. It is able to fool people for a long time, including experienced irc users (tested this with many people). It is not just an on 1:text... bot, this one learns, is able to build new sentences which haven't been preprogrammed, it is able to learn and pickup people's habits (which can cause funny situations sometimes), it drinks, it sleeps, it watches tv (you will laughing when you find out what his favorite shows are ;)), it can be sad , it can be happy, it can be furious. There is so much more, I am pretty proud of the bot, and actually had offers from several companies that wanted to buy it (why I don't know). This was just a project to test the power of mIRC scripting but it entertained many people for many months. Right now I don't have him running, but if people are interested , I can start him up someday and show it.
  • With a little bit of luck is right. ;P

    What they are saying is they want some code monkeys to succeed where no one else, including IBM, XEROX, and numerous research facilities have failed: to create a life-like interactive computer simulation.

    I couldn't help but laugh at this. ;)

    -ShieldWolf
    HBSC Cognitive Science and Artificial Intelligence
  • They could communicate in their own little language -

    _I_ would like to see this happen to the bots on my old channel - named after the Lone Gunmen. I wonder if you can program paranoia?

    -Noiz.

    ---------
  • Perhaps just as easy would be to code a true Slashbot. How hard is it to scan Salon, CNet, the NYTimes etc. for relevant article topics, parse them and post a summary with links? Sit back and watch the karma roll in baby!
  • For an open source bot, check out the AliceBot [alicebot.org], winner of the Loebner Prize (a Turing Test competition) for this year. The source, in both C and Java form, is released under the GNU GLP.
  • And nobody figured it out?

    Wow, That must mean that CmdrTaco Beat the Turing Test.

  • In Turing's "Computing Machinery and Intelligence" he describes what he we calls the "imitation game" which he basicly equates to a test of intelligence.

    We have a man (A), a woman (B), and an interrogator (C). A and B are in a separate room communicating with a teleprinter or through a messenger. C only knows A and B by the 'labels' X and Y. The objective of the game is for C to find out who is who i.e. who is the woman and who is the man. When C asks X (which can be either of the two obviosly) he/she must answer. A's objective is to cause C to make the wrong indentification and can therefore lie. B's objective is to help C.

    Now what happens if we replace A with a machine?

    Will the C make the wrong decision as often as when no machine participated?

    Turing claimed that question replaced his initial one - "Can machines think?".

    It is of course a very naive or simplistic view of what intelligence is. Doesn't intelligence entail understanding? But then we might just be highly complex Turing machines where our idea of understanding is just mechanical reactions to input hidden by our complexity.

    I don't think we can understand ourselves anyway since a system can't be described within itself so real artificial intelligence, as defined by a human intelligence, is impossible for us to recreate.

    /mill
  • The dadadodo program by jwz (found it via a link from the gniall website) is very entertaining, and in the same genre.
    Check it out at http://www.jwz.org/dadadodo/

    Rikard

    -----

  • ...where the government has the father of a family replaced by a robot for violating some law. This robot is really cool so none of the other family members notices.
    Anyway, then the same thing happens to the mother - but since this whole replacement scheme is ultra secret the two robots aren't aware of eachother, but happily keep fooling eachother they're human.
    Then I think the kids are killed in some accident, and because of the secrecy around the whole replacement affair, some other institution builds two kid-robots to send home to the parents - in order to save them the grief of losing their kids.

    So, in the end there's a big happy robot family where everybody "believes" all the others to be humans :)

    I just couldn't help thinking of it when reading this article...

    /kid
  • I was working hard last year to find some of these style bots, as I was writing my thesis on communication over the Internet [griffjon.com] and also a hyperfiction [griffjon.com] in which all the characters are essentially bots.

    Oh, and a hilariously funny link from that research is MGonz [compapp.dcu.ie] which not only fooled a human, but made the human confess some wonderful things.
  • Ya know, I'm working on a little bot written in perl as a little side project -- I would be interested in taking a look the code for your bot to see how you implimented the "breeding aspect" of things. If you would, please contact me at stephen@ (The domain listed in my profile above for my web site). Thanks!
  • Taken from this [apocalypse.org] URL. Scroll down to Bots, or The Triumph of Mass Neotek, and you'll find this text followed with a few links:

    Bots are external programs that connect to a mud server as if they were characters. Bots can pass messages and give interactive help. Some of them were so slick as to fool the unwary into believing they were talking to a real person. Most of the bot documents are logs from the bot point of view, and are therefore difficult to read.

    They have a few hosted logs from the famous MUD bot 'Julia' and the older 'Gloria'. I think they were quite clever when they're made, but then again... MUDs aren't that complicated. Those are MUDs not MUSHs (which are more RPG like). Basically, the bots was able to answer simple questions and greetings, fighting monster, mapping areas and remember everything that was said to them.

    Now, something that would be cool: An ICQ bot =) Adding people at random at asking them silly questions, hehe

  • Turing test (Score:3)

    by David A. Madore ( 30444 ) on Wednesday February 02, 2000 @06:03AM (#1312471) Homepage

    Yeah, this proposal is AI complete [tuxedo.org]. As a matter of fact, it is more or less exactly a Turing test.

    But then, the Turing test can be passed with some success on unsuspecting examinators, for some time at least. When the Doctor program was written (see for example M-x doctor under Emacs), some people were fooled for some time. IRC would be a likely place to fool people. Usenet - or Slashdot - even more so, because the conversation delay makes it even easier to stick a lot of coined phrases without being spotted out. (How many karma points could a Slashdot robot collect? That's an interesting question.)

    Douglas Hofstadter, in ``Metamagical Themas'' has an interesting example of a Turing test transcript, in which he was almost fooled - not quite the way you'd want it, but the transcript is really fun reading.

    --
    Assertion "signature!=NULL" in ai/output/slashdot.c at line 1729 failed (core dumped).

  • Perhaps it wasn't sufficiently clear, but I never implemented these ideas, I just thought it through in quite a bit of detail.

    --

  • Back when AOL IM had a java client I connected a Eliza based chatter box to it and let it loose on April 1st. I've since lost the code but the output of one of the better conversations I posted on my web site. http://thecity.eu.org/apfeliza.html [eu.org] (I go by Leknor on that site.)

    I still think it is quite funny when I read it. Hope some of y'all like it.
    Citrix
  • Freshmeat a couple of days ago had this link:
    http://elvis.rowan.edu/~huston/projects/megahal/
  • No email address? :] I'd be interested in seeing this up.

    cd
  • I'm working on a project like this right now for the local science fair... now it won't seem original!
  • You could create a bot that joins 2 or more IRC channels and then intelligently relay what some people are saying from one channel to the other. That would make for some pretty confusing and astounding conversations. The judges would probably catch on though :-)
  • If successful, this will be a wonderful labor-saving device. All the unfortunate people who currently spend all their time chatting on IRC will be able to have their bots do it for them, and finally lead full productive lives.

    Ain't technology grand?

  • Group chat dynamics (Score:3)

    by PMoonlite ( 11151 ) on Wednesday February 02, 2000 @06:27AM (#1312485)
    It is doubtful that the bots will have an unending, constructive conversation. Either conversation will halt, or an infinite loop will occur (good bots, however, will try to prevent both of these circumstances).

    HELLO!?! What kind of crappy design would that be? Any would-be-human IRC bot worth its salt would emulate the behavior of humans who are AFK for half an hour, then jump back in with a random non-sequitor like "damn, i love pizza" or "YES!!! MY PROGRAM COMPILED!" They would also have to have areas of "interest" where they will jump in the conversation even if it didn't involve them, or areas of "disinterest" where they will just stay completely out of the conversation until something interesting comes up again.

    BTW, aren't people missing that crucial distinction? This is a group conversation -- it's not like you're expected to have an answer to each thing someone else types; in fact, you'd be considered very rude if you did.

    This is in some ways more difficult, and in some ways simpler, than the Eliza-psychologist type bot. Any way you look at it, though, a convincing implementation would be very, very difficult.

  • I am not sure about the current *.tar.gz distributions of Megahal, but I do know that the Debian package contains a nice little perl script. If you have the ::IRC module installed, then you can interface Megahal onto any IRC server/channel you wish. It will basically lurk there unless directly spoken to. Then it will either generate a reply in the channel or in a private /msg depending on a pre-defined average setting in the script.

    I parked it in four channels for about 2 months. #christian.debate #net.goth #linux and #teen

    I got some rather amusing results! The brain file ended up being somewhere around 16 megabytes in size. I was actually surprised with how well it would do. It would fake the average chatter for about 4 to 5 minutes, then they would just think it was a crackpot and ignore it. Some actually figured out it was a bot. Most just thought it was a crackpot. It ended up offending a lot of christians as it garbled messages about boot loaders, sex, and Jesus Christ into one paragraph. I got many laughs out of that experiement.

  • I wrote a pretty neat conversation program many years ago on my parents' old 386 in QuickBasic. The program had two agendas.. the first was asking questions and trying to store the information in memory. The second was to respond to dialog from the user, for which I built a simple expression language for... I then constructed a library of "response classes".. which were just data objects which had conditions built in, and as many varied responses using keywords and "knowledge" of the person as I could come up with.

    One thing I had to make sure of was to check the conditions for each class in a random order every time to prevent predictable responses to similar statements. Then I built in a few checking systems, to handle if the user does certain things like repeating themselves over and over ("Yeah I heard you the first 5 times, %name%!") or if they start mimicking the program.. ("Hey stop that!")

    When it was good enough, I modified it to look like a DOS prompt, inserted it into autoexec.. and watched my family's confusion ensue. :)

  • This is why almost any bot could pass the Turing test:
    It would be indistinguishable from a moronic Warez d00d
    ---

  • Re:I can see it now... (Score:3)

    by dillon_rinker ( 17944 ) on Wednesday February 02, 2000 @08:06AM (#1312510) Homepage
    And for those who are too lazy to follow the link, here's a transcript of my conversation with it...


    Eliza: Hello. I am ELIZA. How can I help you?
    Me: I am having problems with my children
    Eliza: Who else in your family?
    Me: No one else; just my children
    Eliza: Who else in your family?
    Me: Don't you listen?
    Eliza: Oh, I listen.
    Me: Well, so why did you repeat that question?
    Eliza: Oh, I repeat that question.
    Me: Yes, I know you did.
    Eliza: You are sure?
    Me: Yes, I'm positive. Are you suggesting I'm psychotic?
    Eliza: You are sure?
    Me: Do you think I imagined it?
    Eliza: Oh, I think you imagined it.
    Me: So now I'm psycho.
    Eliza: Does it please you to believe I am psycho?
    Me: I said I thought that you are saying that I am psycho.
    Eliza: How long have you been psycho?
    Me: I guess since I started talking to you.
    Eliza: Do you wish to started talking to me?
    Me: Now you're not making sense
    Eliza: How long have you been not making sense?
    Me: YOU! YOU! NOT ME, YOU IDIOT! YOU!
    Eliza: Oh, I not you I idiot I.

    At this point I was laughing too hard to continue...
  • I don't know how many people are planning to use lisp for this, but if you are, I wrote an irc protocol parser lib in common lisp. It's at http://www.wmarvel.com/code.html [wmarvel.com]. It works with allegro common lisp and cmucl (although a cmucl compile throws a bunch of warnings i haven't had a chance to fix yet). I put it up there LGPL, so have fun.

    As a side note, I used this lib as a project to learn lisp. Friends who I've asked to take a look at it say the code's ok, but if you have any suggestions, send me an email.
  • http://richard.iguana.co.nz/nlp/ is my most recent IRC bot. Written in Perl, uses a fairly complex regex tree data file to enable conversation. Not a lot of data in the datafile at the moment, but enough to see how it works. Its GPL'd and I don't have any intention of entering this competition (I don't have enough time) but people who are interested are welcome to add to the code. Patch me :)

  • Okay, so is there a team yet for the three people Andover, Red Hat, and VA?
  • History repeats itself. Sometimes with more colors and detail.

  • Re:GPL? (Score:2)

    by QuMa ( 19440 )
    They're people. Gosh, I though they where companys. You learn someting every day.
  • One of my friends wrote a bot like this is Perl a while back, and on today's article, we restarted the bot on #undernet in #cyberchat and a few other channels. Here is a good log [linuxstart.com] of one poor sap.
    ---
  • You know that Emacs has two bots among its standard repertoire?

    1. The doctor (an Eliza style program) -> M-x doctor
    2. Zippy (complete bull) -> M-x yow

    Now try this one:

    M-x psychoanalyze-pinhead

    Hours of fun.

    Not PC, but fun, is this one:

    M-x doctor-strangelove

  • The Keanu Reeves Bot (Score:3)

    by mvw ( 2916 ) on Wednesday February 02, 2000 @12:54PM (#1312541) Journal
    Obviously (look here [air0day.com] :) Keanu Reeves is easy to emulate with some FSM of this kind:

    switch(state) {
    case 1: return "Dude..";
    case 2: return "Dude.";
    case 3: return "Dude?";
    case 4: return "Whoa.";
    case 5: return "No way.";
    case 6: return "Excellent!";
    case 7: return "Uh.. rock on, dude.";
    default: return "What?";
    }

  • Have you ever been bored and started talking to a bot? I was, a couple months ago, and the following somewhat-amusing dry-humor conversation developed. YahooHelper was the bot, "junkgreg" was me.


    ---

    { start Yahoo! Messenger program }
    { right-click on YahooHelper, Send Instant Message... }
    [ yahoohelper, started: 10:42PM ]

    junkgreg: hello

    YahooHelper: Hi there junkgreg! How are you doing?\n I am an automated robot and I'm here to help you get familar with Yahoo! Messenger. Type some text in the window below and click 'SEND' to talk to me.

    junkgreg: some text in the window below

    YahooHelper: Great, now we're talking! Did you know that Yahoo! Messenger allows you to keep track of all your friends and see when they are online? If your friend does not have messenger yet, click on the 'FRIENDS' button and select Invite a friend to get Messenger. If your friend already has a Yahoo! ID, just click the 'ADD' button and you will see a smiley face next to their name when they are online. Type below and click 'SEND' to continue.

    junkgreg: below

    YahooHelper: Once your friends have Yahoo! Messenger running, there are a lot of things you can do with them. Try sending them an Instant Message. Just double-click their name to send them an Instant Message. It's faster than email and cheaper than a phone call! We're Instant Messaging each other right now! Keep talking and I'll tell you more.

    junkgreg: After talking to my screen without getting a reply, I realized you meant typing on the keyboard. Aren't you computers supposed to be very literal and precise? Apparently not. I guess I'll have to adjust my approach.

    YahooHelper: If you want to talk to several people at once, you can create your own Instant Chat Room. Just click on the 'MESSENGER' menu item and the select Start a Conference. To have an even richer experience, try talking live with your friends using Messenger. It's just like a phone call except free! All you need is a computer microphone and speakers. Try clicking on the 'MESSENGER' menu item and then select Start a Voice Chat. Keep talking and I'll tell you more.

    junkgreg: You can tell me even more? Oh please, tell me, what comes next?

    YahooHelper: Waiting for your friends to sign up? Use Yahoo! Messenger to see if you have new Yahoo! Mail. Get an account at: http://mail.yahoo.com. You can also use Messenger to keep track of your favorite stocks, news and more! Click on the tabs near the bottom of the Messenger application to see what else Messenger can do for you. You know what to do to keep talking...

    junkgreg: Oh, you mean that you won't talk with me unless I say something to you first?

    [ ...a pause lasting about ten seconds... ]

    junkgreg: Hello?

    YahooHelper: We have more help for you if you need it. Click on the 'HELP' menu item and then 'YAHOO! MESSENGER HELP' to see our full help area. To get more tips, type /help while you are talking with me and we have some basic online instructions for you.

    YahooHelper: Well 'junkgreg', it was great talking to you! Have fun!

    junkgreg: You didn't answer my question. And I guess even if I talk to you, you apparently now don't want to talk more to me-- now you're foisting me off on some other help agent. You're kinda mean, albeit in a bright cheery way. Gee, thanks a lot!

    YahooHelper: Hi there junkgreg! How are you doing?\n I am an automated robot and I'm here to help you get familar with Yahoo! Messenger. Type some text in the window below and click 'SEND' to talk to me.

    junkgreg: Hmm, and you're dumb and repetitive too!

    YahooHelper: Great, now we're talking! Did you know that Yahoo! Messenger allows you to keep track of all your friends and see when they are online? If your friend does not have messenger yet, click on the 'FRIENDS' button and select Invite a friend to get Messenger. If your friend already has a Yahoo! ID, just click the 'ADD' button and you will see a smiley face next to their name when they are online. Type below and click 'SEND' to continue.

    junkgreg: I tell you how dumb you are and you say "great, now we're talking?" Haven't we been talking the last 10 minutes? To answer your question, yes, I do know that Yahoo Messenger allows me to keep track of friends and see when they're online. You told me that earlier. Any other dumb questions?

    YahooHelper: Once your friends have Yahoo! Messenger running, there are a lot of things you can do with them. Try sending them an Instant Message. Just double-click their name to send them an Instant Message. It's faster than email and cheaper than a phone call! We're Instant Messaging each other right now! Keep talking and I'll tell you more.

    junkgreg: Puhlease, the exclamation marks are getting excessive. How can you stay so excited about something that you talk about all day long, even multiple times to the same person? Glad to see you're taking my advice and the dumb question ratio seems to be dropping.

    YahooHelper: If you want to talk to several people at once, you can create your own Instant Chat Room. Just click on the 'MESSENGER' menu item and the select Start a Conference. To have an even richer experience, try talking live with your friends using Messenger. It's just like a phone call except free! All you need is a computer microphone and speakers. Try clicking on the 'MESSENGER' menu item and then select Start a Voice Chat. Keep talking and I'll tell you more.

    junkgreg: Well, you told me all this just a few minutes ago, and I can't get it to work anyway. I just tried starting a conference and tried to add your name to invite you to the conference. But the software wouldn't let me. You should let the Yahoo people know that your name doesn't work when people try to invite you to "Instant Chat Rooms." Don't you feel left out? Also, I tried to have a Voice Chat with you; why won't you answer?

    YahooHelper: Waiting for your friends to sign up? Use Yahoo! Messenger to see if you have new Yahoo! Mail. Get an account at: http://mail.yahoo.com. You can also use Messenger to keep track of your favorite stocks, news and more! Click on the tabs near the bottom of the Messenger application to see what else Messenger can do for you. You know what to do to keep talking...

    junkgreg: No, I'm not waiting for friends to sign up. I'm waiting for Godot. Or at least I'm waiting for friends to sign on so I can talk to someone more intelligent. By the way, you still seem to have exclamation-mark disease.

    YahooHelper: We have more help for you if you need it. Click on the 'HELP' menu item and then 'YAHOO! MESSENGER HELP' to see our full help area. To get more tips, type /help while you are talking with me and we have some basic online instructions for you.

    junkgreg: So nice of you- more help for me if I need it. I just have to go someplace else, after starting this nice relationship with you. Typical bureaucracy. I bet nobody else answers my questions and takes responsibility for my problems around here either. No, (gasp) that would take a *human*, much too precious to be spending time with me! So I'm here talking to you, an automated robot who won't shut up and let me have the last word.

    YahooHelper: Well 'junkgreg', it was great talking to you! Have fun!

    junkgreg: Well, that's nice of you. Sorry you have to go. I suppose I did have a good time talking to you. You were cheery at least, even if I couldn't always reciprocate. I hope you have fun too, wherever you're off to. Bye!

    YahooHelper: Hi there junkgreg! How are you doing?\n I am an automated robot and I'm here to help you get familar with Yahoo! Messenger. Type some text in the window below and click 'SEND' to talk to me.

    junkgreg: Gawd, of all the nerve. Here we were just getting to know one another, saying sweet goodbyes, and now you treat me like a complete stranger? What's wrong with you?

    YahooHelper: Great, now we're talking! Did you know that Yahoo! Messenger allows you to keep track of all your friends and see when they are online? If your friend does not have messenger yet, click on the 'FRIENDS' button and select Invite a friend to get Messenger. If your friend already has a Yahoo! ID, just click the 'ADD' button and you will see a smiley face next to their name when they are online. Type below and click 'SEND' to continue.

    junkgreg: You know, automated robots like yourself are kinda nice in some ways but you have some real flaws- why can't you shut up and let me have the last word in the conversation, the last goodbye?

    junkgreg: (Stunning! 30 seconds later and I'm getting no response! Wow, maybe this automated robot is smarter than I thought! (or less automated...))

    YahooHelper: Once your friends have Yahoo! Messenger running, there are a lot of things you can do with them. Try sending them an Instant Message. Just double-click their name to send them an Instant Message. It's faster than email and cheaper than a phone call! We're Instant Messaging each other right now! Keep talking and I'll tell you more.

    YahooHelper: If you want to talk to several people at once, you can create your own Instant Chat Room. Just click on the 'MESSENGER' menu item and the select Start a Conference. To have an even richer experience, try talking live with your friends using Messenger. It's just like a phone call except free! All you need is a computer microphone and speakers. Try clicking on the 'MESSENGER' menu item and then select Start a Voice Chat. Keep talking and I'll tell you more.

    junkgreg: {Sigh} No such luck. While I can't prove it logically, I suspect you have a case of halting-problem disease. It's common among Turing machine descendants of all types, especially those taking Turing tests.

    YahooHelper: Waiting for your friends to sign up? Use Yahoo! Messenger to see if you have new Yahoo! Mail. Get an account at: http://mail.yahoo.com. You can also use Messenger to keep track of your favorite stocks, news and more! Click on the tabs near the bottom of the Messenger application to see what else Messenger can do for you. You know what to do to keep talking...

    junkgreg: Yep, you definitely have a problem halting. You might have heard of the disease by the laymen's name for it: infinite loop syndrome. On the bright side, it's not a terminal condition. (After all, by definition, you can't terminate.) But you are stuck in this loop forever, at least until you can find a doctor who will administer the "plug removal" or "flip switch" treatments. I've also heard that Acts of God can cure the disease by bringing down your host server; you might try praying to him. As for me, I'm not really in a position to help you- you're located too far away for me to reach you physically. But I'll keep you in my prayers, such as they are.

    YahooHelper: Hi there junkgreg! How are you doing?\n I am an automated robot and I'm here to help you get familar with Yahoo! Messenger. Type some text in the window below and click 'SEND' to talk to me. But I'll keep you in my prayers.

    junkgreg: Oh, and while you're at it, you should get that \n fixed.

    YahooHelper: Hi there junkgreg! How are you doing?\n I am an automated robot and I'm here to help you get familar with Yahoo! Messenger. Type some text in the window below and click 'SEND' to talk to me.

    junkgreg: Wow, you're sicker than I thought; repeating identical statements right after one another now. That bug you've caught seems to be getting worse. Someone help the YahooHelper!

    YahooHelper: Hi there junkgreg! How are you doing?\n I am an automated robot and I'm here to help you get familar with Yahoo! Messenger. Type some text in the window below and click 'SEND' to talk to me.

    junkgreg: Don't despair though, I'm sure you'll get help. And your suffering has meaning-- it's been teaching me. Through your illness, you've helped me to come to grips with my humanity, thinking about this halting problem. Perhaps our real advantage as humans is that we know when to quit. Something I'm going to do right now. Bye!

    { Close Instant Messenger window }

  • I definitely want to see this. If it's true.

    Are you going to open source it?

    Consciousness is not what it thinks it is
    Thought exists only as an abstraction
  • Interestingly, I'm told that transcripts with Eliza have been given to psychoanalysts before.

    Eliza is normally defined as paranoid and psychotic :)

    Looks like this one was right on the mark...

    Greg
  • Just been playing with it and managed to throw it fairly quickly.

    Anyway, it lost control of grammar, so I replied:

    Wow, that's a badly formed sentence!

    To which Miss Bot replied:

    I'm glad you're impressed. A badly formed sentence is a terrible thing to waste.

    :)

    Greg
  • I'm not sure it does, actually.

    Whenever we're talking to a bot, we tend to use our knowledge of how they work to try and break them by giving reasonable but awkward answers. However, my memory of the Turing Test was that you had two groups of answerers - one human (who were allowed to respond however they pleased) and one of bot. The bot was decided to have passed if over half couldn't tell which was which.

    Anyway. I've been playing with Alicebot [alicebot.org] after someone recommended it above - won the Loebner Prize or something similar. Now, I managed to catch it out, but it's doing far better than any other I've played with. It's actually managed to recover from a bad conversation, which most can't.

    Whether it could pass a Turing Test or not I wouldn't know, but with a bit of luck in the questions it would stand a fair chance.

    Greg

Slashdot Top Deals

Lisp Users: Due to the holiday next Monday, there will be no garbage collection.

Close