Slashdot Mirror
Kubrick's AI Spawns Distributed Client / Cognition
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."
"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.
I only post comments when someone on the internet is wrong.
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.
Welcome to the A.I. game from the maintainer of the Trail. Please check out the Cloudmakers website--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)
http://danhon.com/
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
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.
At the moment, the Slashdot minds are focused on a puzzle: Which link points at a page which states what the puzzle is.
I'm one of the moderators for Cloudmakers. Those looking to get involved should definitely start at the Cloudmakers home page.
We're always looking for new players to jump on board, so read the FAQ and subscribe!
<humor>Oh, and buy CM T-shirts too!
When I moderate, I only use "-1, Overrated". That way, I never get meta-moderated!
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 and Guide to get up to speed. If you want to try to play the game yourself without spoilers, check out the Journey. And if you're really into it after that, join our mailing lists which are linked on the main page of cloudmakers.org.
---
Brian Seitz (praying to the slashdot effect gods)
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)
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
People see the world as they are, not as it is.
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.
I don't need large brains to have a good time.
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.
I don't need large brains to have a good time.
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/
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.
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
"It is seldom that liberty of any kind is lost all at once." -David Hume
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.
Eh...
If you want to have some fun with this bot, 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 on picture-rate.com
"and dear god does this website suck now." -- CmdrTaco
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 comic strip
"It is a greater offense to steal men's labor, than their clothes"
Question...
What does getting a story acccepted do for you?
"Beware of he who would deny you access to information, for in his heart, he dreams himself your master."
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.
"Beware of he who would deny you access to information, for in his heart, he dreams himself your master."
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.
"Beware of he who would deny you access to information, for in his heart, he dreams himself your master."
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.
"Beware of he who would deny you access to information, for in his heart, he dreams himself your master."
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.
"Beware of he who would deny you access to information, for in his heart, he dreams himself your master."
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.
"Beware of he who would deny you access to information, for in his heart, he dreams himself your master."
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?
--------
Bleah! Heh heh heh... BLEAH BLEAH!!! Ha ha ha ha...
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.
--
--Got Lists? | Top 95 Star Wars Line
Well, here's one for you. I opened up two copies of the AI page 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.
It may look like I'm doing nothing, but I'm actively waiting for my problems to go away.
--Scott Adams
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.
This is not a Fugazi
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. 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