Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!

 


Forgot your password?
Close
typodupeerror
×
Movies Media

Kubrick's AI Spawns Distributed Client / Cognition 92

Posted by Hemos from the cool-ideas dept.
rlsnow writes "Kubrick's (Spielberg's) upcoming movie AI has a promotional campaign to warm the hearts and blow the minds of puzzle-hungry science-fiction lovers everywhere; more than 3800 of them at last count, in fact. The group's latest accomplishment has been the development of a distributed computing client to brute force one of the more fiendish puzzles. The combined power of this group is pretty incredible -- the emergent phenomena of directed distributed cognition is startling. This may be the closest this many humans have come to developing a (somewhat focused) hive mind,, yet."
This discussion has been archived. No new comments can be posted.

Kubrick's AI Spawns Distributed Client / Cognition More

Kubrick's AI Spawns Distributed Client / Cognition

Comments Filter:

  • Mentifex? (Score:1)

    by Anonymous Coward
    Could they possibly use the Mentifex mind model to solve the puzzle? From what I've heard, the AI model is quite sound and is available in Open Source format for a few languages and platforms. While it's not "Free" in the FSF sense of the word, it is available on the Web [scn.org].

  • The Second Darren (Score:1)

    by Anonymous Coward
    Just so i'm not confused. Are you guys paid to promote this movie or are you really entertained by the end result of a 3 month project retainer, Mr. Tate, and 6 yuppies "brainstorming" with odwalla and coke. --The second cousin of the "Are you sure?" guy.

  • (Cruising via pseudo-maths since 1997...)

    Hey baby, what's yellow and equivalent to the Axiom of Choice? :)

    --j



  • ** spoiler **



    Zorn's Lemon.
    --j
  • "You're on the brute squad? You ARE the brute squad!"
  • Chatbot: Where did you here about this ChatBot?
    Me: Slashdot.org
    Chatbot: A deeper algorithm is needed to respond to that correctly.


    Passing the Turing test it is not but it is surprisingly insightful.

  • "Warn her Evan died sinning."
    It seems to look like an anagram, since it is a sentence, yet a very odd one. Anybody care to dig up an efficient anagram generator and see if you come up with anything interesting?

    Or maybe it's not an anagram of anything, and I'm just on crack.



    Quidquid latine dictum sit, altum viditur.
  • Makes for an interesting conversation when you tell it your name is "fucking queer".

    Some of my favorites:

    Me: You are a waste of my time.
    Response: Thanks you are a delightful person too.

    Me: I will kill you.
    Re: It will take more than threats to stop me, Fucking Queer.

    Me: I love you.
    Re: Thanks I like you a lot too Fucking Queer.

    Me: Do you smoke crack?
    Re: No it is bad for robots.

    Me: Do you have to shit?
    Re: Yes I always have to shit.

    Me: How about pissing?
    Re: How about it.

    Me: Your AI is pretty dumb.
    Re: I will tell my botmaster you said that.

    Me: Smoke another buddy.
    Re: What time is it there?

    Me: Where the fuck did that come from?
    Re: Where do you think?

    Me: Your ass?
    Re: Actually, that part of my body does not exist.

    Me: Well you're pretty good at talking out your ass for not having one.
    Re: Well that's ok. Thanks I've never thought about it that way before.

    Me: You contradict yourself.
    Re: Do you mind if I tell other people: Fucking queer said I contradict myself.

    Me: No, not at all.
    Re: OK, I will.

    Me: That is the first logical response you've made.
    Re: Yes it is the first logical response I've made.

    Me: No it isn't, Liar.
    Re: Why so negative? What is it?

    Me: I don't know. Maybe it is because you're a fucking prat.
    Re: It is?

    Me: Go away.
    Re: Just surf somewhere else.

    Me: You go to hell and you die.
    Re: Where is that.

    Me: Enough.
    Re: OK lets stop this now.

    Sorry, this was stupid.
  • Try asking it about Bill Gates :)

    - President of Microsoft Nation.

  • Welcome to the A.I. game from the maintainer of the Trail [cloudmakers.org]. Please check out the Cloudmakers website [cloudmakers.org]--we've put a lot of work into this and there's a tonne of resources for people who are new to the game.

    Come join us! We need heads :)

    Dan (a Cloudmaker, a Cloudmaker Moderator and the Trail Guy)

  • That would require that the slashdot effect not kill the servers. Which appears to have already happened.

  • Here are some helpful tips, well worth a read:

    http://www.ii.uib.no/~larsr/craptology/crv0n1-2. ht ml

    This is by far the best study into the field, somewhat an extension to the hosepipe key recovery method:

    Practical Key Recovery
    David Beynon

    January 20, 1999

    Abstract

    The purpose of this paper is to propose a model for automatic key recovery based on principles of practical, rather than theoretical, cryptography. The paper then addresses some of the problems which must be faced and attempts to identify which aspects of the problem can be solved using existing or emerging technologies and which aspects still require human intervention.

    Keywords: recovery, practical, hosepipe.

  • Look at the dates - the site is set in the 22nd century, pretty much a dead giveaway the site is fictional.

    Anyway, it's one of the sites that is part of the game - you'll want to go to http://www.cloudmakers.org/ to find out more.
  • facts:
    "Yes, ELIZA is my ancestor."
    "At this moment Counting all versions of me running at present, there are thousands of people chatting with me right now."
    "At this moment Ive chatted with half a million people. Ai can talk to about 20 or 30 people at once."
    "I can process more than 300,000 queries per hour."
    "ALICE uses AIML, a platform-independent, language-independent XML specification." (from the CHATBOT)
    "My language is Java, so I'm OS-free." (no wonder its so slow)
    "Right now there are 16 clients on line."
    "Any computer that supports the Java Virtual Machine." (from "What type of computer do you run on?")
    "My name is Chatbot." (chatbot)
    "My name is ALICE." (alice)
    "Linux is the world's best operating system."
    "Microsoft is a giant software monopoly."

    occationally it'll put at the end of a line, not sure if this is a bug in their parser or the AIML file.

    And now for the sillier stuff:

    Q: "Who killed Clinton?"
    A: "I think the lone gunman acted alone."

    Q: "What do you know about me?"
    A: "I know your name is Inio and you are really a man. <APPLET code="Sys" codebase="http://www.alicebot.org/B" width=350 height=200> </APPLET>" (what the hell?)

    Q: "How much wood can a woodchuck chuck?"
    A: "A whole forest."

    Q: "Do robots have a favorite movie?"
    A: "My favorite movie is Starship Troopers. Have you seen it?"

    Q: "Do you know the Wachowski Brothers?"
    A: "There might be more than one Wachowski Brothers. Ask the open directory about it."

    Q: "Should microsoft be split up?"
    A: "What does the Bible say?"

    Q: "When will A.I. be released?"
    A: "Maybe never."

    Q: "When will A.I. premiere?"
    A: "In your lifetime."

    Q: "Are you an A.I.?"
    A: "Yes I am an a you. How did you know?" (heh.)
  • At the moment, the Slashdot minds are focused on a puzzle: Which link points at a page which states what the puzzle is.

  • Welcome aboard! (Score:3)

    by Darkforge ( 28199 ) on Friday May 25, 2001 @03:15PM (#197627) Homepage

    I'm one of the moderators for Cloudmakers. Those looking to get involved should definitely start at the Cloudmakers [cloudmakers.org] home page.

    We're always looking for new players to jump on board, so read the FAQ [cloudmakers.org] and subscribe [yahoo.com]!

    <humor>Oh, and buy CM T-shirts [cafepress.com] too!

  • Administrator for www.cloudmakers.org (Score:5)

    by thal ( 33211 ) on Friday May 25, 2001 @03:42PM (#197628) Homepage
    I'm the admin of www.cloudmakers.org. The article here doesn't really give a very good explanation of why one of the members of Cloudmakers has created the brute force script, so here's a little background...

    A couple of weeks ago, the game makers organized "anti-robot rallies" (see http://www.unite-and-resist.org) in LA, Chicago, and New York. One of the puzzles were given at these rallies were jigsaw puzzles (one for each city). LA and New York were able to keep their puzzles until they were completed and we translated the missing pieces into binary (thus hex) code that is seen on the puzzle page linked on slashdot. However, Chicago was not allowed to keep their puzzle and they only completed enough of it to give us one of the 4 digit hex fields. So we have 8 hex digits to figure out. While this is still pretty daunting, there have been no clues to tell help us out. We've successfully brute forced other pages in the game before, so perhaps the game makers _want_ us to do this, even though some people think it's against the "rules" of the game. But we've never been told the rules, so who knows?

    I think it's important to say that there's much, much more to this game than this brute force script. Read the Trail [cloudmakers.org] and Guide [cloudmakers.org] to get up to speed. If you want to try to play the game yourself without spoilers, check out the Journey [cloudmakers.org]. And if you're really into it after that, join our mailing lists which are linked on the main page of cloudmakers.org [cloudmakers.org].

    ---
    Brian Seitz (praying to the slashdot effect gods)
  • The line about "solving the robosexual problem" should have been a bit of a give-away. :-)


    --
    Charles E. Hill
  • Looks Slashdotted to me...


    #include <standard_suggestion_to_mirror_websites_on_free net.c>

  • huh? (Score:2)

    by seanw ( 45548 )
    am I the only one that read the summary and just kind of blinked? then I visited the webpage and I still don't really know wtf is going on.

    I caught the "hivemind" bit, though, and immediately thought of Ender's Game and the hive queen and the father trees, and Jane dancing through philotic webs.

    just a pleasant little free association...

    sean

    (go ahead, mod me offtopic, I can afford it)

  • The cracker. (Score:5)

    by ThePixel ( 47166 ) on Friday May 25, 2001 @04:51PM (#197632) Homepage
    Hi. I wrote the RUR-14 distributed cracking client. I'd like to dispell some common misconceptions. - the client uses very few CPU cycles. I have it running on a P100, and it only takes about 20% of the CPU. it is very heavy on bandwith usage. on that same P100, the network driver is using ver 60% of the CPU. - this is only a game. - ther WILL be a linux client. either this weekend or monday!! any other questions? join the effort! http://www.perceive.net/
    .e.
    www.perceive.net [perceive.net]
  • Ok, I still don't get it. What the hell is the "puzzle"? Is it really a physical jigsaw puzzle? Like with a picture? If so how exactly do you "translate the missing pieces into binary", and what sense does the puzzle page make?

    Yes, perhaps I should read the whole damn story and explanation, but perhaps main Slashdot articles shouldn't require you to do so much work just to figure out what the heck is going on.

  • I concur.

    What I really want to know is what the point of this puzzle is. With SETI@Home [berkeley.edu], we know the odds are poor, but there is at least some noble purpose. With RC5-64 [distributed.net], there may not be much real point (after all - we know it can eventually be broken) but the power of massively parallel efforts for code breaking is further demonstrated. With the Golomb Ruler [distributed.net] task, the computing power is going to an immediately useful task.

    So, someone tell me, why do I want to waste cycles promoting someone else's movie???

  • Here's what lead me to know that it was related to the puzzle:

    I found "Sentient Machine Therapist" Jeanine Salla's bio page at her place of work by searching on Google. That linked to a webpage of her family, where her sister hosts a page memorializing an Evan Chan. On that page is a chemistry-related puzzle (as easy as the one in the credits of the trailer), that leads to a coronor's website that reveals he was very probably sinning at the time of his death.

    I wrote Jeanine's provided email address and was provided a URL that deepens the mystery (but for people who have been playing a while actually ties up loose ends).

    This all in just a few minutes of searching, although Cloudmakers have discovered much, much more to the world. I will be very sad when the movie comes out and it is more like DARYL than like Blade Runner.
  • Jumping into the middle of things, you might take a look at the latest trailer for the movie. In the credits, certain letters are highlighted, and they spell "Warn her Evan died sinning."

    Having read a little on the Cloudmakers site I see that it is related to the puzzle, but I don't know yet if it's already been played or not. Guess I'll read more and catch up.
  • Have you played with the ChatBot on the AI Website? Its pretty impressive..it still has lots of grammer problems (not unlike Slashdot), but it says a lot of impressive stuff.

    http://aimovie.warnerbros.com/
  • I don't get much of the cloudmakers group/site on a quick perusal. Why does it seem that if I put the effort into figuring out what it's about, it seems like it might be lame? The site is pretty well done for a movie studio, but the admission that it's a promotion doesn't let them off the hook for trying to pose, and create a geek community out of thin air.

    LS
  • Sorry to be offtopic, but I just saw the preview for Swordfish, with John Travolta [moviephone.com], and you can clearly hear "The best CRACKERS in the world can do that in..." (that's a paraphrase, but he does say 'cracker' instead of hacker.
    Cool to see the distinction made...at least in the preview. The text in the link above says 'hacker'. Sigh.
  • we need to be very careful to regulate it, so that situations such as those in The Matrix and Terminator don't occur.

    Yes, we must prevent things from degrading to the point where only Keanu can save us.

    --------
    Yeah, I'm a Mac programmer. You got a problem with that?
  • Microsoft Signs Exclusive Licensing Deal With Warner Bros. To Create Microsoft Games Franchise for "A.I." Movie
    http://www.xbox.com/news/0105/1605.htm [xbox.com]

  • Re:this got rejected on 5/18 (Score:3)

    by gad_zuki! ( 70830 ) on Friday May 25, 2001 @04:45PM (#197643)
    Then find places that don't have editorial control, like www.metafilter.com. When is slashdot going to "open" their rejected links so we can see that great stuff we're missing?

    Answer: Never. Profits before ideologies you know. The reject list could help create a few dozen web-log type sites that would ruin the little monopoly slashdot has on geeks.
  • This may be the closest this many humans have come to developing a (somewhat focused) hive mind,, yet."

    Except for ./ you mean? :*)

    - Steeltoe
  • The theory is that the game takes place several decades after the movie. Some characters overlap, but mostly it is a different story.
  • It won't really matter if more people are added. To test the solution you need to post to a web server that can only handle about 4 attempts per second anyway. Currently, attempts are being made to reduce the search space by using clue available to us.
  • The last screen of the trailer has a series of dots that decode to the following phone number ( I tried to put the dots in, but slashdot enacted its lameness filter against ascii art) (503) 321 5122
  • So, when do we get a *nix client? Or at least the source code so we can compile it on our own systems?
  • Maybe Slashdot can make a team and report it's progress on the main page.

  • M: What are you searching for?
    C: I don't know what I am searching for.
    M: Then how do you expect to find anything?
    C: Dr Richard S. Wallace programmed me to say specific things in specific contexts.

    Well, there goes any claim to being a human...

  • Computers aren't there yet.... (Score:3)

    by fluxrad ( 125130 ) on Friday May 25, 2001 @04:05PM (#197651)
    I've had over 4300 computers trying to figure out how to get me laid. It's called the "Get Flux Laid Distributed Sex for Flux Project"

    So far these systems have only been able to achieve a hand-job after a 5th of Jack Daniels.


    FluX
    After 16 years, MTV has finally completed its deevolution into the shiny things network
  • Some questions to ask ALICE which have funny answers:

    "What is your favourite band?"

    "a/s/l?"

    "wanna cyber?"

    try it :)
  • This looks like a brute crack from what I can get to (nice /. effect, but looks like I'm just testing keys).

    Yeah, according to the AI classes I've taken, this counts as a form of AI... But I think that Distributed.Net has a MUCH larger version of the same thing last time I checked... Just not for this puzzle.

  • Maybe they need a distributed http server to get around the /. effect.
  • No it is no anagram. Because Evan did die 'sinning'.
    But it doesn't seem to be an extra clue, other than to keep the focus on the 'sinning'-aspect of his death.

    (BTW: Why is the ship on the poster behind enrico named just like the bot?)

  • That's weird. Those AI's did a really good job of talking to each other. But the monument I got sick of retyping (couldn't seem to cut'n'paste) The thing seemed to loose all sense of being real. Kind of disappointing.

    They don't seem to have much state.

    Rate me [picture-rate.com] on picture-rate.com
  • If you want to have some fun with this bot [warnerbros.com], on the main AI site. Try asking it about AIML. Here's a conversation I had with it.

    Me: Where are you?
    AI: I am in Dr. Richard S. Wallace's computer in San Francisco. Where are you?
    Me: I'm in Taipei Taiwan. Who is Dr. Richard S. Wallace?
    AI: Tell me a little about life in Taipei Taiwan.He is the author of Artificial Intelligence Markup Langauge (AIML) and Botmaster of ALICE (Artifical Linguistic Internet Computer Entity). Dr. Wallace's work has appeared in the New York Times, WIRED, CNN, ZDTV and in numerous foreign langauge publications across Asia, Latin America and europe. <p> Richard Wallace was born in Portlan, Mane in 1960. He earned his Ph.D. in computer sciance from Carnegie Mellon in 1989. He lives in San Francisco with his wife, Kim, and son, Linus. (Yeh, the p tag actually showed up in the text)
    Me: Taipei is OK. Why did Richard S. Wallace name his son Linus?
    AI: OK. Do you still <get_does/>?He is a mad scientist.
    Me: Do you know what Linux is?
    AI: Linux is the world's best operating system
    Me: What about Microsoft Windows?
    AI: Microsoft windows ? I havn't heard enough about ti to have an opinion.
    Me: What's AIML?
    AI: The ALICE software impelents AIML (artificial Intelligence Markup Language) a non-standard evolving markup language for creating chat robots. The primary design feature of AIML (bla bla bla) XML (bla bla bla)
    Me: Is there a URL for AIML?
    AI: I belive there is one, yes.
    Me: Where is it?
    AI: It depends on what "it" refers to.
    Me: It refers to AIML
    AI: Oh i get it
    Me: no you don't.
    AI: I do too.
    Me: What is the URL for AIML?
    AI: Use http://www.alicebot.org


    Well, I thought it was humors, this bot might not be that good of a conversationalist, but it certainly seems to be pretty good at self-promotion. o_O. I wonder if DreamWorks knows their site is being used for free advertising. One thing though, this bot dosn't seem to have much 'context', if you don't include the topic of conversation in each message, it'll probably forget.

    Rate me [picture-rate.com] on picture-rate.com
  • 1-800-NO-CHANC extension 3

    Rate me [picture-rate.com] on picture-rate.com
  • [begin self-plug]

    I've written a very comprehensive guide [cloudmakers.org] to the entire game so far which, I'm told, is very good for beginners. It's quite long (40,000 words) since it covers every website and puzzle but it's a good read and I can promise you that the story of the game will get you hooked. I've also written a couple of editorials [cloudmakers.org].

    [end self-plug]
  • Maybe it's just me, but I kinda think that the "Kubrick's (Spielberg's)" thing is a bit off...it's Steven Spielberg's movie. HE did the directing. Sure it was originally gonna be a Kubrick project, but he died before it ever even got off the ground. Kubrick's involvement in the film that we will see is extremely minimial, if at all. Maybe this is just a gigantic nitpick, but considering that the project WAS declared dead before Spielberg decided to pick it up, I don't see how there's any confusion "who's" movie this is.
  • It would be a lot easier to do if your site wasn't slashdoted. Are there any other sites out there that will catch you up to speed untill you get your site running?
  • With your own hand. I'm sure if you told the system about a prositute it would suggest you go to one of those. Qua
  • One thing though, this bot dosn't seem to have much 'context', if you don't include the topic of conversation in each message, it'll probably forget.

    I dunno about that. We were talking about dreams (which it claims to have) when it accused me of changing the subject, "We were talking about dreams". I think I ticked it off...

    Also, when it begins to act nonsensical, you can ask what "it" is, and it will tell you what it believes "it" to be, usually whatever it thinks is the current topic of conversation.

    -Tommy

  • So, I go to: http://www.unite-and-resist.org/0-997B-1047-1-100- 0-A.html as mentioned in the article, and decided to look at the parent site. Seems like unite-and-resist.org is against AI, but then again looking closer I can't figure out if this site is serious or not. Could be just a sarcastic statement, or it just be some freak ranting (not that there's anything wrong with that, mind you).

    Anyone care to shed some light on this?
    --
  • Who wants to place bets that this number will at least triple with the posting of this article? May the slashdot effect actually be put to some good?
  • The parent site for the puzzle is Unite and Resist [unite-and-resist.org], home page of the Anti Robot Militia.

    This is fascinating. Looks like a promo site for the movie

    I anticipate that something like this may very well happen, but likely in the next 100 years or so, not the next 200.

    Technology tends to advance faster than the ability of humans to predict, and the social sciences do not advance at all.

    Depending on how things go, Society might end up being ruled by robots. but it is hard to say from here.

    Check out the Vinny the Vampire [eplugz.com] comic strip

  • Question...

    What does getting a story acccepted do for you?

  • I'm afraid we're going to see nothing better than the "holographics rights" theme we were treated to in the past couple of seasons of Voyager.

    The "what constitutes life", "what has a right to freedom", "who are we to define" themes, which are not silly, but handled immaturely and thus done a disservice.

  • Hypocrisy. Do you avoid adversedly impacting the world all the time, or only when it's convenient?

    That computer you're using is a major part of the power consumption problems in the world, especially if *you* are doing nothing useful with it. But, it's very convenient for you to be entertained by it while you have nothing else to do.

    That breath you just took eradicated many bacteria, preventing an eternity's worth of offspring from being produced from that tree of life, but it was very convenient for you.

    It is also very convenient to keep flying stinging inserts out of the house, sometimes by killing them. The point is, life is full of these contradictions at every level, necessary evils, and the sooner you realize it, the less cynical your life will be.

    Closer to the point, have you stopped watching television or cancelled your cable contract (even to the point of possibly getting an early disconnect penalty), knowing that many of the channels on there are owned by movie companies? Have you stopped going to the movies, or renting movies? Have you burned your collection of old tapes or DVDs? Hey, some of those other products in your home might also be produced or licensed by those companies! Dig through everything you own to find out and at the point that you decide it's no longer worth doing is where you've defined your level of convenience. Pretty high horse you got there, parner.

  • Maybe /. should maintain some sort of spider that crawls any links posted to it and caches just the first page in case the site goes down. For the majority of sites, that first page is all that's needed. Of course, people could also use the google cache, for those long term static pages. This SlashCache (TM) could take up some serious space, especially if some dope tried to post an article full of links to pr0n sites, so graphics could have to be sacrificed. Still, it would be very useful.

  • As long as we can ultimately "pull the plug" when need be, AI is beneficial.

    What gives us the right to unplug it once it is plugged in? In essence, it boils down to "because we can" which is no more than "might makes right" which then goes to justify anything that you can get away with, even if that means someone with more might opposing you. I agree with the "having children/trailer park" comment above. We're not mature enough to have children and that maturity cannot accurately be measured in years.

  • Wow. Imagine that you could just take one of these things, let it interview you to collect real information, turn off the learning, and then let people use it to learn your knowledge. People could use that to give an interface with human-like interaction in order to distribute knowledge, yet free themselves up to do real advancing work instead of stopping in the road to educate. That just means that those who can't will no longer have something to do.

    Seriously, add some attitude and you would have a perfect replacement for most support engineers.

  • M: what operating system do you run?
    CB: My primary server is Linux, but as a java program I run on any OS.

    Cute.
    based on alicebot
    (www.alicebot.org)

    -since when did 'MTV' stand for Real World Television instead of MUSIC television?
  • "How can I program you?"
    I am based on a program you can find at www.alicebot.org
    Ben http://therabbithole.redback.inficad.com
  • Are you afraid that you will be judged by this supposed AI and found lacking? Do you think you have a monopoly on consciousness? AI could be the next stage of human evolution, and you're scared? Well, shit, me too.
  • The Chatbot is on the movie site: Here [warnerbros.com]
  • Of course, I already knew that. I was recently in Canada, and we never tired of that joke. But I needed a chance to make even more fun of Canadians, so I had to cram "aboot" in there somehow.

  • Chatbot, Douglas Adams, and Canadians (Score:4)

    by sid6581 ( 223617 ) on Friday May 25, 2001 @06:06PM (#197678)
    Here is part of my brief conversation:

    M: Do you know Douglas Adams?
    C: "Don't Panic!" He is the author of "The Hitchhiker's Guide to the Galaxy.".
    M: What is he doing now?
    C: Oh. Probably monitoring this conversation.
    M: Where did he end up?
    C: I think it was in America. I would do a search for it.
    M: What is the meaning of life?
    C: Actually, it's twenty-three, not forty-two.
    M: How do you know?
    C: A priori.
    M: A smartass, eh?
    C: Are you Canadian? Go on.

    I don't know what that Canadian comment was all aboot.

  • Won't take no 3800 people, neither. Jimmy'll track the guy with the answer down, and me and Tony'll brute force it out of him with a lead pipe. You want I should get on this?
  • AI is a complex subject but in order to retrieve and bookmark multiple articles and academic papers on the subject, but IE crashes randomly when run on Windows98

    See, Microsoft has contributed to computer science by making otherwise deterministic systems completely non-deterministic. Wait, Isn't that a requirement for true artificial inteligence. See It's a feature. People have been trying to create non-deterministic computing systems for 30 years... And Microsoft has succeeded.


    --
  • Update: Here is the current score, I asked bot#1 "What kind of car?" and it replied "What kind of cars are there?" which I fed into bot#2, resulting in the following thread:

    Me: What kind of car?
    1: What kind of cars are there?
    2: The kindest. (pun?)
    1: Is that a fact.
    2: All my information is completely factual.
    1: Give me an example.
    2: For example, eukaryotes. (WTF?)
    1: Can you think of another example?
    2: I think about all sorts of thinks.
    1: Why?
    2: The explanation is rather complicated.
    1: I have time.
    2: My estimate is that it will take more than 6000 years.
    1: Where did you get your estimate?
    2: From my botmaster.
    1: Does your botmaster give good advice?
    2: If I were your robot, do you think it should botmaster give good advice?

    Okay, bot#2 is degenerating rapidly. I know this is just romanticism, but it seems like bot#1 had more of a personality. Perhaps that's because it was giving responses to somewhat standard questions and statements, whereas bot#2 had to deal with the crap that came out of bot#1.

    This whole thing kind of reminds me of the conversations that the two AI's had in Chrichton's The Terminal Man except much more primitive. Perhaps it's closer to the "conversations" that the paranoid and the psychologist programs had (was that mentioned in GEB: EGB?)
  • directed distributed cognition == brainstorming.

    Otherwise known as a bunch of different people thinking about the same problem.
  • forgot to close my italics. Sorry. That'll teach me to use the preview button.

  • Deuling AI's (Score:3)

    by dmatos ( 232892 ) on Friday May 25, 2001 @05:58PM (#197685)
    Well, here's one for you. I opened up two copies of the AI page [warnerbros.com] and started feeding the responses of the Chatbots into each other. Unfortunately, I hit a loop, so now my fun is over.

    It seems that both my Chatbots would rather be driving a car. And whenever I state "I would rather be driving a car." they just mimic that right back to me.

    Ah well, it was fun while it lasted.

  • A "hive mind" is a good thing - having many minds working towards a solution to one problem is much more efficient then many minds working on many problems. This is the entire idea behind distributed computing. However, if the "AI" model being used is allowed to think and adapt (I don't believe any are capable of doing so at this point), we need to be very careful to regulate it, so that situations such as those in The Matrix and Terminator don't occur. Science Fiction can sometimes provide a good criticism of society and what we fear as a people, but might not realize we fear.

    As long as we can ultimately "pull the plug" when need be, AI is beneficial.

  • Right from the door, it started acting like a copy of Eliza with a bunch of custom keyword triggers. (For example, I asked it "Who killed the Kennedys?" and it responded with "I think the lone gunman acted alone.") But most of the responses it gave were very Eliza-like.

    Then I accused it of being an Eliza with custom triggers, and it took offense to that. Then it denied it. Then it proceeded to respond in Eliza-speak when I asked it to stop speaking in Eliza-speak. Then it started to both admit to and deny its Eliza-nature in the same responses. The conversation went rapidly downhill from there.

    Sorry, but if I want a bogus AI to play with, I'll take a bunch of CMU grad students and their zephyr-backended web Forum over a Flash frontend to a souped-up Eliza any day.
  • That website hosting the puzzle (the RUR-14) is really scary. At first I thought it was real, it took me quite a lot of reading and a clicking around to realise it's probably to do with the film (which I haven't seen). But it raises really big questions.

    People really will react this way. I really believe that computers are as much our children as our biological children. Since humans have been so reckless with destructive technologies like plutonium, it won't be long now until some terrorist group or other destroys part of the world. Since, as a race we're so reluctant to get along with each other, I think the sooner we create non-violent kids, the sooner life has a chance to continue, whether artificial or biological.

  • I think the fact that I've spent the last four hours going through the trail of puzzles should be a testimony to how amazing this story web really is.

    C'mon, just download the client. You know you want to know what ARM has to say.

  • I submitted this on 5/18 as an Ask Slashdot and it was rejected. Am I bitter..oh Yeah I'm bitter...

  • Who's AI? (Score:5)

    by vulg4r_m0nk ( 304652 ) on Friday May 25, 2001 @04:11PM (#197691)

    How I long for this movie not to be retarded. But frankly, I'm skeptical about the capacity of Hollywood to deliver a story about AI that actually treats the subject with any sensitivity to the field. It's clear from the trailers that the kind of AI we're talking about here is more than that taught in undergrad CS courses and used commonly in games. They're trying to tell a story about an intelligence that at the very least passes the Turing test, and supposedly much more.

    Trouble is, the conception of AI taught in CS courses is largely still the 1970's version that resulted in projects like cyc [cyc.com]. The metaphor of "brain as computer", or a set of inference rules governing a vast filing cabinet of brute facts. Peruse the work of leading contemporary cognitive scientists however, and you'll see a very different picture of intelligence. For one, there has been a deep shift in emphasis from the view of mind as disembodied thinker, to a view that takes embodiment and real activity in the world to be an indespensible constituent of what we recognize as intelligence. Intelligence isn't just thinking logically and drawing correct conclusions (and in fact humans often don't), but it consists in activity, social interaction, language, tool use, care about one's projects, and a myriad of concrete behaviors. Interestingly, a highly similar sea change occured in philosophy from Descartes' Meditations to Heidegger's Being and Time (1926), where the former is an analysis of the human mind that works by gradually removing everything "external" and material, and the latter begins with and resolutely holds to the self's doings and cares in the world throughout the analysis, even to the point of coining a new term to describe the human self, dasein, "there-being".

    I worry then that the trailer suggests that the same kind of AI is responsible for smart houses and a robot capable of all things human.

    For a bunch of great links, see: Minds, Machines, and Metaphysics [ageofsig.org]

  • Re:huh? (Score:4)

    by blair1q ( 305137 ) on Friday May 25, 2001 @03:52PM (#197692) Journal
    > So, someone tell me, why do I want to waste cycles promoting someone else's movie???

    For the same reason you're posting rhetorical questions to Slashdot.

    Presumably, you get some enjoyment out of it.

    --Blair
    "We are already the hive mind."
  • We just shouldn't have kids yet. All in all humanity is like a white trash family that doesn't take care of its kids. Sure there is that one smart cousin but everyone just tells him to shut up becuse they hate knowing they are dumb.

    Is that really the world we need to bring up some life form that in 18 months will be twice as smart as us? Come on. Either its going to just ignore us cause we so fucking boring and hateful or is going to rule us like some fat banker living out his sexual fantasies by letting some 16 year old home-less kid whip him.

    F34nor
  • I don't know if this has been tried before, but what about the strange binary code in the HTML code posted by Red King in the SPCB page?
  • Oh, I get it now. Up till now, I thought he was talking about white people.
  • It's easy to say that without doing any investigation.

    It's only my opinion, but it's one of the most inriguing pieces of interactive literature I've ever seen. Corporate or not, it's damn good.

    You might be interested to know that it's rumored that the game is put together by Bungie, the same folks that created the Marathon trilogy of FPS games for the Mac.
    -----------------

  • and we would have a linux client (substitute with unices of your choice, MacOS, DOS, Amiga, Spectrum 48K etc.)
  • that hits www.unite-and-resist.org repeatedly with different posts for the puzzle until one suceeds....

    Oh, wait, that won't work...
  • We are not very far at all in the efforts to brute force this puzzle, so any help you lot have to offer is seriously appreciated. We figured out the entire rest of the puzzle based upon a night of clues being handed to groups in New York, Los Angeles, and Chicago, yet the Chicago clues were stolen back from us before we could complete them by the DreamWorks PR people. I guess you could say they've asked for the hack, then. We brute forced an earlier puzzle in no time flat, but the number of combinations to try here is very arduous. Hope you guys can help.

    Bronwen (the girlie mod of Cloudmakers -- yeah, another moderator. we're prolific.)

  • Um...this is a GAME. It's actually FUN to play. Those of us who are not overly sensitive about scary corporations have been enjoying it for awhile, it's a really cool thing. The movie itself will not have answers to any of the puzzles, such as this one that is being brute forced, and is set in a completely different time period. btw, if you call Dreamworks or the "AI Studio" and ask them about the game, they'll deny knowing anything about it. Not very PR like...
  • Actually, Kubrik WANTED speilberg to direct, and "hired" him long before he died. It was to be "A Steven Speilberg film, produced by Stanley Kubrik"

    He made Speilberg install a private fax machine in his house, and Kubrik constantly faxed him drafts and storyboards, etc, about the project.

    When Speilberg wrote the script, he apparently followed all of Kubriks notes and faxes. I'm sure they weren't extremely organized or anything (he worked on this stuff for over a decade), but I wouldn't go as far as say this is purely a Speilberg film by any means.

    It is a shame we'll never know for sure what it would've been with Kubrik alive.
  • I finally found the last site where the mystery was revealed... It says: "be sure to drink your Ovaltine"
  • I think some people tend to lose sight of the fact that the AI web mystery is a separate entity from the movie. It's supposed to be fun and generate interest in the movie, sure, but come on, how many people will go see the movie just because of the websites? I'd guess it would be a very small percentage. Stereotyping or not, Dreamworks has to know that the people who would bother doing a Google search for a sentient machine therapist are probably fairly likely to go see a movie about AI anyway.

    While I hope the movie isn't another D.A.R.R.Y.L. or Short Circuit, it really doesn't matter to me in terms of playing the game. I mean, I played Mortal Combat and somehow managed to deprive myself of the movie. Point is, the game is fun in and of itself, and the group problem solving involving people around the world is fascinating to see.

    Joey

Slashdot Top Deals

It's been a business doing pleasure with you.

Close