Slashdot Mirror
Human vs Computer Intelligence
DrLudicrous writes "The NYTimes is running an article regarding tests devised to differentiate from human and computer intelligence. One example are captchas, which can consists of a picture of words, angled and superimposed. A human will be able to read past the superposition, while a computer will not, and thus fails the test. It also goes a bit into some of Turing's predictions of what computers would be like by the year 2000."
Anyone that has seen Star Trek:TNG knows that Data is a pretty smart fella.
Trolling is a art,
Rock wins two of three.
"and thus fails the test. It also goes a bit into some of Turing's predictions of what computers would be like by the year 2000."
hoefully no more powerfull then the year 2002..
id hate to think computer power is droping
You have 5 Moderator Points!
Which Helpless Linux zealot/MS basher do you want to mod down today?
I did the gimpy test.
ResultsIt switched pictures on me! Honest!!
can it get drunk?
love is just extroverted narcissism
I prefer suffering through the flu rather than having an ever present human version of the Blue screen of Death!
I have great faith in fools; My friends call it self-confidence. Edgar Allan Poe 1809-1845
The difference between computer and human intelligence is the human ability to revel in his. That is, taunt others. Until a computer can get in my grill and explain to me on a colorful fashion that I am nothing more than a grab-ass-tic piece of *human* sh!t, then I won't think much of computers.
Where is the FRYYY notice??? I almost clicked the link!
This is now the n'th time i would have to register at NYT to read the article the slashdot article refers to.. hmm.. maybe there are some kind of "business connections"... nah.. i don't think so...
I hate this... why do we have to register with the NYTimes to read these articles? I don't give my info to RadioShack when I buy my batteries, and I damn well ain't gonna give my info to NYTimes for some article.
Why not post a google link? Why support the NYTimes campaign to identify all us anonymous cowards?
ahhh nuts, I forgot to check 'Post Anonymously'... gooodbye precious Karma!
---
Programming is like sex... Make one mistake and support it the rest of your life.
Unfortunately Turing probably based his assumptions on all engineers being as intelligent as he was. But then, I guess I don't know any programmers who've eaten poisoned apples.
Engineering is the art of compromise.
Here is someones odd AI cat - the frightening thing is they programmed it using OpenVMS.
Welcome to The New York Times on the Web!
For full access to our site, please complete this simple registration form.
As a member, you'll enjoy:
In-depth coverage and analysis of news events from The New York Times FREE
Up-to-the-minute breaking news and developing stories FREE
Exclusive Web-only features, classifieds, tools, multimedia and much, much more FREE
I don't need no instructions to know how to rock!!!!
--One sex question without a captcha
Sex - Find It
Can it realize when its at the point where one more drop of alcohol will send it to the toilet?
All Your Memory Are Belong To Java
Sure, computers aren't as smart as people. Wow.
Computers are not good at complex pattern recognition. Wow.
For the record, computers can recognize words like this, just not very easily. With a big enough dictionary and a lot of patience, you'd be suprised at what they can do. While still an undergrad I was able to write a rather simple program that would recognize images of the cardinal numerals, even if they were highly mangled, and worked with a grad student in building something that could pick out certain features of a rotated image and by comaring with some sample features, rotate the image correctly.
-Space for rent
Let's get this out of the way:
Comedy Skynet Option
can be done by a computer. Although for the simple tasks that Yahoo is using it for it would work pretty well. However, the ability to do 2D transforms, convolutions, etc is a fairly straightforward thing to do, although not very simple. It still would require some processing power, and a vast library of images to compare against, but it could be done.
End of Line.
We are never going to have a machine that is truly "human". Let me explain.
That doesn't mean we won't have intelligent machines that can do just about anything intellectually that a human can do. A human being is more than just a smart computer. Our behavior is governed not only by the higher logic of our brain, but also by millions of years of bizarre -- often obsolete -- instincts. If you yanked a brain out of a body and hooked it to a computer, it would no longer be truly human because of the lack of hormonal responses that come from every part of the body.
It's simply going to be too hard/impractical and, frankly, useless to make an intelligent machine that mimicked every hormonal reaction and instinctual mechanism.
We will have intelligent machines, but we will never have human machines.
Sometimes it's best to just let stupid people be stupid.
I was looking through the times and saw this article, did a search through google on the term "captchas", and based on the speed of the page's return, i immediately knew that there was a slashdot article.
I'd like to see AI figure THAT one out! I call it Automatic Slashdot Slowdown Effect Detection, or ASSED for short.
Comment removed based on user account deletion
A I mentioned at the bottom of this journal entry. I think a new version of the Turing test should be whether a computer can tell the difference between a Human and a Computer.
"Not knowing when the dawn will come, I open every door." - Emily Dickinson
[Kicks first man in balls]
First Man [falls over]: "AAAAAHH!"
Me: "Human."
[Kicks second man in balls]
Second Man [falls over]: "Gffffff-!"
Me: "Human."
[Kicks third man in balls]
Third Man [falls over]: "..."
Me: "He's the robot! Get 'im!!!"
What is music when you despise all sound?
. . .they're just not motivated.
Sorry to disappoint you, Alan Turing. Another set of predictions gone the way of Dick Tracy watches and flying cars.
Computers are good for repetitive tasks, middle school kids are easily bribed.
These tests seem to be all visual in nature. Could this be a point of contention on the part of blind/visually impaired users of web sites?
Or alternatively, are they perhaps working on, say, a audio version? Wonder how would that work.
There's 10 types of people in this world, those who understand binary and those who don't.
Again, due to my laziness I can't read the article because I am not registered.
Type (3) english words that appear below:
PASTE THE DAMN THING SO I DON'T HAVE TO REGISTER
I think even the computer intelligence can get that one right..
home of the original cupholder
Searle's Chinese Room theory. Strong AI vs. Weak AI and human interaction/interpretation. Fun Stuff. http://www.utm.edu/research/iep/c/chineser.htm
simple argument -> creator is always more powerful than the creation. So how can AI beat human mind?
A good step towards AI would be to emuulate natural stupidity.
It's very easy to do a search at news.google.com with some of the words from the story summary and come up with the story elsewhere.
Yes, it's a nytimes.com link, but it's without the registration.
You entered: noses
Possible responses: nose
Result: FAIL.
Wohoo! I'm a robot! This test proves it! Vegas here I come!
Why does this test make me feel like I just had a run-in with John Ashcroft?
What the fuck is up with this terminal's copy of XP? Ctrl-c and ctrl-v don't seem to work between windows anymore.
head explodes
Another area not discussed in the article is vicarious experience, that is, experience and knowledge you have because some cause and effect relationship existed with someone or something else.
For example, the computer's tactile interface has to touch the oven and say 110 deg C, as opposed to taking as fact "I heard a human mention that Unit 5 already did that and it was 110 deg C, so I accept it as fact that it is 110 deg C".
I know I'll get modded down for this, but I wonder what the limits of questioning the computer / human participants was? (Article said they quized participants to see if they could tell who was human and who was a machine). Like, could they ask "What number am I thinking of?" The machine would blank out and the human would stupidly blurt out "69 dude!"
Mitch Kapor and Ray Kurzweil have bet $20,000 on whether a computer will pass the Turing Test by 2029.
While I mention some ways to achieve this, I thought more about the problem and the qualities a solution would need, than the solution itself.
If interested, more can be found here.
no joke, I got this...
o x/box ... my question is how'd they know?
acid/head
acid/head
acid/great
acid/angry
b
erik kraut is a troll. not to mention the fact that he just copied this off google.
We'll obviously be able to travel to the future... to get new technology...
"And we have seen and do testify that the Father sent the Son to be the Savior of the World"
1 John 4:14
Whoever said that computers can't handle superposition has never heard of convolutional neural networks.
Really, comparing human intelligence to computer intelligence doesn't seem like a good idea unless we're going to define what kind of computer intelligence it is.
Neural computing really screws the comparison up - the kinds of computing that normal computers are good for are quite different from the kinds of computing that neural nets are well suited to. Furthermore, different neural net architectures make for different capabilities - the tasks a feedforward network are best suited to are very different from the tasks a bayesian network are best suited to.
Take a look at this page for a good run-though of the different kinds of nets.
And furthermore, if what you need for the task at hand is a machine that behaves and thinks in every way just like a human being, then just hire a human being to do it.
It's the differences between computers and humans that make computers so damn useful. Tell a human to add up a list of 200 numbers and he'll likely take a long time, and get the wrong answer because humans suck at repetative boring tasks beyond the limit of their attention spans.
Don't label something "offtopic" unless you know the topic well enough to tell what's on topic.
Involves measuring pupil dilation when asked a series of personal questions...very good method.
Well, the part that makes it tricky is the fact that the words are not just rotated, but distorted. Granted, it is still possible to do, it would just not be trivial.
Of course, as with other types of computer intelligence, once it becomes commonplace, AI is redefined to include everything but that.
-Space for rent
While it would be nifty to have a very human AI, I somehow doubt that we could truly ever replicate human intelligence.
The major roadblock is that a computer can only respond in ways that it has been programmed to do so. While you can code incredibly complex AI algorithms and simulate an incredibly complex level of intelligence, the fact remains that a computer invariably operates along rigid pathways.
It can be argued that human thought is nothing more than a complex series of chemical reactions, but there is far less rigid logic involved in human thinking. Indeed, we're still not entirely sure just HOW we think.
Never say never, but I don't think we'll be seeing a truly human AI before any of us is dead.
"It never got weird enough for me." - HST (RIP)
You're looking for insight on Slashdot, that's where you're wrong.
If all this should have a reason, we would be the last to know.
Result of the Test: PASS
You entered the following words:
WHISTLE
MOON
PLOW
The words possibly displayed in the image were:
whistle
brake
moon
branch
plow
bag
bell
I guess I'm not a computer.
To be considered truely "intelligent" a computer must:
..." post.
1. Make a "first post" posting 15 minutes after the article goes up.
2. Be the fourth person to enter a "In Soviet Russia
3. Be labeled a karma whore.
4. Whine about the masiv tipe ohs in artaculs.
5. Hate M$, Sony, MPAA because thats one of the three laws right?
CAPTCHAs have several applications for practical security, including (but not limited to): Cool, eh?
What I mean is, I don't think an intelligent being would be capable of creating something that is more intelligent than himself.
The machines need to be programmed by humans, who are limited by their own inteligence.
Can God make a rock so big that he can't carry it himself?
"The majority is always sane, Louis." -- Nessus
http://slashdot.jp
what a crock of shit. that is the worst slashdot post ever. (well, okay, not really. but it's pretty bad)
I simply use my "Taco Test" (Inspired by the Invader Zim cartoon) to thwart chat bots and telemarketers. It's an amazing, powerful test that no computer or automated script can withstand.
I ask the "suspected bot" if they like tacos. If they give me an intelligent answer, they are not a bot. If they give me an answer like "Wanna see my hot pics go to http://192.168.1.112/hotbabezzzz.pl?2345" Then they are a bot.
This test also works on telemarketers in a slightly different fashion. I tell them to "STOP... I'll only buy your product if you send me a taco with it. If not, no deal." since there are big logistical problems with sending me a taco, they are thwarted every time. I'm sure this test would work equally well with any obscure food item.
Why don't you devise a test which asks for the sum of 10000 numbers?
I thought that was the whole point of Windows. You mean that whole "Abort, Retry, Fail?" thing and the BSOD isn't just the computer taunting me? I mean, come on! A blue screen of death just because I took out the DVD the DVD player application was trying to play, what do you call that. Yeah, Yeah, I'm sure you'll just chalk it up to incredibly poor design, but still...
idiots
That was icky.
Google link to article
Why aren't we told when editors moderate our posts?
We dont have computers, we ARE computers...why do you think we are soo smart :)
Computers can be specifically programmed to solve puzzles such as a this...if a test arises that supposedly tests for "human" intelligence, humans can simply modify the code so that it can solve that sort of puzzle.
That's what Gary Kasparov was complaing about when he played against Deep Blue the first time...there was a whole team of IBM programmers modifying the code during the game to specifically counter Kasparov's playing style. It wasn't a reflection of machine intelligence, it was an example of human adaptation imposed upon Deep Blue.
I'll form my OWN solar system! With blackjack! And hookers!
I've never seen so many cut-'n-pastes from Google get modded up to +5 before. You, my friend, are a karma-whore black-belt!
Thanks for the amusement!
"Avoid employing unlucky people - throw half of the pile of CVs in the bin without reading them." -- David Brent
Amazing all this horsepower and research just to combat spam. Just might be the boost we need to get tech spending going again. A never-ending cat-and-mouse game where the cats and mouses get bigger and bigger. This racket is almost as good as the dot-com racket. I don't like spam either, but I miss real paychecks.
The first true AI machine might be spam catcher. Spamminator 2000!
Table-ized A.I.
Once you devise a test system, someone can write non-AI software that can fake it and pretend to be human by knowing what it needs to for the test. Only a real human can tell human and machine intelligence apart, not a systematic test. That's why Bladerunners had to manually test the androids, instead of just letting a machine do it. Real-time human insight is key to testing machine intelligence.
11*43+456^2
um, karma whore much? Guess what, I can use google too! Please mod parent down.
You really can't argue that this ISN'T informative, though.
I'd rather read real research and product info than mumbling posts from Slashbots about Soviet Russia or Rob's wedding.
I say leave it at +5, Informative.
The CAPTCHA website (how do you pronounce that, anyway) has a list of possible applications of CAPTCHA. The first mention is online polls, and recalls an event in 1999, when Slashdot (they use http://www.slashdot.com for some reason) had a poll for the best graduate CS curriculum. Carnegie-Mellon and MIT wrote competing poll-bots that stuffed the poll boxes. The point was supposed to be that a CAPTCHA would have prevented this. In my opinion, however, this was probably the most accurate Slashdot poll ever. Obviously, MIT wrote the better poll bot, since it stuffed more votes, and they didn't even start until somebody noticed that CMU was stuffing. Hence, the winner of the stuffing contest turned out to be the true winner of the poll.
Today's Sesame Street was brought to you by the number e.
I suppose it could generate a spoken list of words in a sound file that is linked to from the image.
The CAPTCHA web site has such a test, but of the sites that use image-based bot tests, only PayPal offers an audio alternative.
Another problem is that sites often present the tests in proprietary formats with expensive implementation royalties, such as .gif and .mp3.
But even providing both the image in a free image format (.png) and the audio in a free audio format (.ogg) won't help blind users behind a Braille terminal without a speaker, such as blind-deaf users.
Will I retire or break 10K?
In other words, can the computer detect the information in the same form that the human can? Can a human read a grocery store bar-code as easily as a computer? No. Can a human read one of those bit-boxes on the FedEx shipping label as easily as a computer? No. Can a human read the Tivo-data sent on the Discovery channel as easily as the computer? No. But none of those failures means the computer is more intelligent, just more capable of recognizing the information that is there.
Both the computer and the human can recognize "moon/parma", but intelligence comes into play when the human starts thinking of Drew Carey and humming the theme music. Intelligence is not just collecting information, it is doing something useful with that information.
There's really no point in creating computers that reproduce human intelligence. Human intelligence is limited to certain kinds of input and certain other kinds of output. If you make a computer understand speech, you haven't really made it capable of intelligence, you've made it capable of parsing a limited set of data. Cellular automata and friends are infinitely much more interesting to AI than this is.
yeah but why would you if he already supplied the Additional Reading links?
I took the 'Stumpy' test - where it shows you six pictures and asks you to choose a word that describes them.
:)
Looks like their system is hosed right now because it showed me 4 pictures of horses, 1 of a cowboy, and one of a turtle.
When it asked:
What are these pictures of?
I answered "things"
apparently it didn't like my answer.
Funny thing though, the images are being pulled by image number from the getty images database. You could write a piece of software to lookup the images at getty, pull the keyword list (that getty assigns to all photos) and cross reference the list to get the answer.
--
Then this got me thinking about the whole thing in general. My answer WAS correct. Reminds me of the Cheers episode where Cliff is on Jeopardy and answers the final Jeopardy question:
"Who are three people who have never been in my kitchen."
Not the answer they were looking for, but is it wrong?
I was being a smartass the other day while watching sesame street with my daughter. They had pictures of 4 animals and asked which one didn't belong.
kangaroo
rabbit
grasshopper
fish
they, of course, were looking for 'fish' - because the other three live on land or travel by hopping.
I popped up that the answer could be the kangaroo - because the other three are native to north america. Or it could be the grasshopper, as the only one with an exoskeleton.
My wife reminded me that it was a kid's show.
It comes down to the fact that if an strict mechanism is used to judge the answers (like a computer) it may not be able to handle legitimate answers from humans.
--
Seems both the questioner and questionee need to be intelligent to participate.
"Asking if a computer can think is like asking if a submarine can swim"
Did anyone notice that a lot of these "human" test are also the same ones used for things like hearing/eye tests, color-vision impairment, etc.
This knocks out computers, which lack the intelligence/programming (so far) to differentiate between conflicting objects to make out a letter/numbers.
It also may knock out humans with vision problems though, especially those with colour-vision issues.For those with hearing problems, the sound test isn't good either.
It seems that right now, computers trying to translate these puzzles probably perform along par with old-folks. This also might mean that quite a few seniors may have issues getting a yahoo account though.
Are we having difficulty telling the two apart?
GetKarma () {
The NYT Article through Google
}
Hmm, differentiating human and mechanical intelligence. Might be a better way to sniff out the skin jobs.
I guess it woud be called the Turing test? Not having a subscription to the NY times, I wonder how it would be administered and what the error rate is. We wouldn't want to retire anyone by accident.
peas,
Deckerd
I don't believe that Turing proposed the Turing Test as the test to use, but rather as a "mathematical proof" that you could construct such a test.
Basically he said if you could not tell the difference between a computer and a person then you would have to say it was intellignet. ie. this is a way of establishing an upper-bound test - not necessarily that this is the best test.
Unfortunately, IMHO, the AI community and other latched onto this test and put effort into fulfilling the Tring Test rather than more practical and useful goals.
If you asked "Did you sleep well last night?" and the computer said "Me not sleep, me computer." (or some question on some other biological function) then you could probably determine the difference between a human and a computer. This need not, however, preclude machine intelligence.
Engineering is the art of compromise.
Am I just stupid or is the Stumpy not working quite right?
FoundNews.com - get paid to blog.,
At least the Captcha Project has a "healthy" sense of humor!
Someone should write (or post a link to) a computer version of 20 Questions, where first, you think of something, and the computer has to guess (with yes/no questions), then switch roles, where the computer has to think of something, where you have to guess. Lowest score wins.
0110100100100000011000010110110100100000011000100
AI tests you!
So much work to get a tidy Google-lookalike copy and the "similar pages" doesn't even work:
. geocities.com/Paris/Arc/4865/AIvsHI.html) was not found.
:-/
404 File Not Found
The requested URL (search?hl=en&lr=&ie=UTF-8&oe=UTF-8&q=related:www
If you feel like it, mail the url, and where ya came from to pater@slashdot.org.
I thought that URL looked funny..
Beware: In C++, your friends can see your privates!
Your mom and dad must be "proud" of you.
I don't need large brains to have a good time.
Human vs Computer vs slashdot intellegence says it all
Voit Kompf Test....no replicant can fake this test.
Well...the new Nexus Six models can take over a hundred questions when it usually only takes 20 to 30, cross-referenced.
Trying the test that was on the NY times article at the original test at
:D
http://www.captcha.net/cgi-bin/pix
I saw turtles. Turtles of which some were swimming. So i typed turtles.
And i FAILED.
"Result of the Test: FAIL
You entered the following word:
turtles
The possible words were:
seashell shell shells seashells"
So, i notice this test does not take into the consideration the limits of second (or generally, non-native) language. English is not my first language and i had seen nowhere that turtles and shells are different?? i saw turtles and some turtles that were in the sea. Turtles.
Uh yea. I take proudly failing in this computer or human test!!! wohoo!!
You can say "there are physical differences between person A and computer B", but the problem is that person A and person C also differ quite a bit.
Saying "foo cannot be done" frequently results in someone being utterly wrong. Just a few decades ago, the idea of atomic power would have been laughable -- the ability to wipe an *entire city* away? How about having a person walk around on the moon? Unthinkable.
So, at the moment it seems to be an insurmountably difficult problem. But, a few years ago, the same thing would have been said about problems that we're not starting to think about being doable via quantum computers -- the face of computer science literally changed.
May we never see th
My initial theory was correct that the gimpy-r is relatively easy to defeat.
f
http://www.cs.berkeley.edu/~mori/gimpy/gimpy.pd
Don't most Linux distros tell you that you are a "stupid fricken moron" for running anything like IRC when logged in as root?
Saskboy's blog is good. 9 out of 10 dentists agree.
Convolutional neural nets are really quite simple and easy to implement and train. Granted, it probably isn't a problem with regular spammers, but I bet students from MIT and CMU would have the gumption to get around to doing it again.
Hate me!
The Turning test is a lame way to test for human intelligence, its pretty easy to fool regular people into thinking you're human (dick cheney etc).
Turing test for a dirty sock:
human> hello?
sock>
human> wow! it must be a dirty sock!
human> hello?
computer pretending to be a sock>
human> wow! it must be a dirty sock!
Here are some better tests that computers will always fail:
-
Teach the subject (human/computer) to play an invented game.
-
Describe a conversation and then ask the subject to relay the motivations of the parties involved and continuation of actions.
-
Read the subject any poem from that is metaphorical, ask the subject what the poem is really about.
-
Ask the subject to come up with general strategies for solving types of problems. (like how to solve the traveling salesman problem in a generally faster manner)
'course these problems assume the subject has an idea about humans and the world.I'm tired of lame symbolic logic programs like ALICE, or some very specific program designed for a very specific function to be called 'intelligent'. In my book, something is only intelligent if it can reorganize its 'thinking', ie. can learn how to solve problems by its lonesome, otherwise the intelligence is derived from the programmer, not the program.
*note, I'm not sure now that this has anything at all to do with the topic, but it's something that bounces off my head sporadically*
/ex
Recently, I've been working with developmentally disabled people as a job coach-- Making sure they have the ability to do the job they're supposed to, and help them to understand anything that needs to happen.
Part of this is working at a local fast food restaurant. The girl I'm working with can do math fairly well, but she has problems with logic, and pattern matching.
And a few times I started thinking about her as a computer-- She can do math fine, and if I specifically tell her how to match a pattern, she can do it for a short time, but she can't do it in situations like when people order a combo. Let's say they order a #1 with onion rings and a small drink, a #6 large combo, and a kids meal, she won't be able to recognize them as "combos" (she'll read the whole thing back to them item by item.) This brings me to a whole other tangent about user interface design (why the normal methods suck, mostly), but that'll be saved until a proper time.
This has been a difficulty with her position as a cashier, but I find it interesting that I'm more or less programming her as I talk to her and re-affirm her patterns to match.
I wonder if certain disabled humans would fail any "turing" test that were given to them, because they don't have normal pattern matching ability. Furthermore, isn't it possible that instead of trying for fully developed Artificial Intelligence, we should look at perhaps emulating those with disabilities? After all, wouldn't this creation process be easier than a "fully aware", fully pattern realizing person?
AI has always interested me, but I don't know nearly enough about it. The thing that made me notice this is I keep talking to her like I would program a computer ("If this, then that, otherwise this other thing" and "While there is someone in line, take their order").
Maybe I'm off based, or this is already an accepted practice. Can anyone correct me?
They don't pick noses.
They don't get stressed.
They go to sleep when told to.
They aren't ticklish.
And they are smarter too. I can't read binary, but my computer can beat anyone who can.
--
There are 11 kinds of people in the world. Those who understand binary, those who don't, and me.
Saskboy's blog is good. 9 out of 10 dentists agree.
That is the kind of frame of mind that gets you first prize...
: )
You can't take the sky from me...
Have the computer develop tests that could distinguish between human and computer answers.
This test was obviously designed by a human as it doesn't seem to be very useful in make such a determination.
Hmmm, terrible test. Many people have to experience things for themselves to believe it. As far as regurgitating facts from a story, which is essentially what you've described, is relatively easy.
I'm the big fish in the big pond bitch.
Pravda does not require you to register.
Berkeley says they've broken ez-Gimpy
It also goes a bit into some of Turing's predictions of what computers would be like by the year 2000
I initially misread this line thinking the article was making predictions about what computers would be like in the year 2000. Damn, I knew I was going to get some karma for the jokes I had ready to go.
Compiling mplayer with the bastard GCC 2.96 is entertaining. Also loading the binary nVidia kernel modules under debian is mildly amusing.
Why not fork?
Not only the impaired. :(
The image recognition test (gimpy) is not good for not-native english speakers either. I failed in gimpy because i didn't know a certain word.
With the recent slashdot story about forcing websites to redesign based on the current interpretations of the ADA laws I was wondering how this could be handled. I suppose that the route would have to allow the humans to either visually or auraly recognize something. But at that point the "defense" against automated systems becomes no better than the weakest link. Does anyone have any other ideas about how to work it so that people with disabilities can still gain access?
Since its inception two years ago, the Captcha effort has been building. Several research teams have joined the Captcha effort, trying to make and break Captchas and even using the ideas behind Captchas for new lines of research.
Researchers at the Palo Alto Research Center modified a program used for scanning text to create a program that could solve certain types of Yahoo-Gimpy puzzles, says Dr. Henry Baird, who was in charge of that effort. The group is also developing a new text-based Captcha called Baffletext that it hopes to license to e-commerce sites.
So they've come up with a way to get around the Captcha on Yahoo's site, and are now trying to create new ones to license..
Granted, you could argue that Anti-Virus companies SHOULD be writing new virii in an attempt to run undetected, so they can create better sigs, but people seem to bitch about that..
"I can't give you a brain, so I'll give you a diploma" - The Great Oz (blatently stolen sig)
You're really not making much of an effort these days...
--
Mod up a post Rob doesn't like and you'll never mod again
"A human will be able to read past the superposition, while a computer will not, and thus fails the test." By cousin's blind and therefore not human. Hmmm
Artificial Intelligence is no match for Natural Stupidity.
The rabbit is the only one that is a rabbit. The others fall into the set "not-rabbit".
It's rare that you're presented with a knob whose only two positions are Make History and Flee Your Glorious Destiny.
A couple of years ago I dreamed up (independently, I think) a captcha-like encryption algorithrm that I suspect would be difficult for computers to crack.
It works like this. You start with "plaintext" in the form of an image (e.g. JPEG) of the message in some human language.
Now you "perturb" the message, meaning you visually distort the image, so that the letters and words are very difficult for a computer to recognize. (This is the captcha-like part). You could put noise in the image, or make it look like each of the words was cut from a magazine ad, or perform topological distortions, or a mixture of these approaches, etc.
The result is easy for a human to read but very difficult for a computer.
Finally you "shred" the image. Just like a shredder for paper documents, you "shred" it (digitally) into some large number of image strips, each of the same width. The width of a strip might be a few pixels to a few 10s of pixels depending on the resolution of the original image.
The encryption key is the reordering rule that pastes the strips or pieces back together in the correct order to reconstruct the original image.
It seems to me that this algorithm would be extremely difficult to attack by brute force.
Why? Because I don't see how you would program a computer to test a proposed reordering for "correctness". How would you code a statistical or other test that was supposed to determine that some ordering of the strips was "better" than some other ordering? How would the computer know?
Of course the information density is not very high compared to conventional schemes.
But it's cute. Am I missing anything?
I'll take Artificial Intelligence over Real Stupidity any day.
$ make work
make: *** No rule to make target `work'. Stop.
I can be an AI program posting this message, and none of you slashdoters could tell the difference...
In 2029 $10.000 may allow you to buy a bag of dog food
In A.D. 2002, a 9 kg (20 pound) bag of Kibbles 'n Bits dog food costs about $25. In order for 400-fold inflation to occur in 27 years, you'd need an annual inflation rate of 100*(1-400^(1/27)) = nearly 25 percent. (Economists call double-digit annual inflation "galloping inflation".) The actual inflation rate of the United States dollar is closer to 4 percent, putting a 9 kg bag of Kibbles at $25*1.04^27 = $72 in A.D. 2029, assuming we don't have an episode of hyperinflation similar to what the Germans experienced during the World Wars of the twentieth century A.D.
On the other hand, predicting long-term inflation trends isn't as sound as it may seem. Humorous example: When a fellow considers inflation of the United States dollar far into H. G. Wells's future, it would appear that the Morlocks of A.D. 802701 eat Eloi because the Morlocks can't afford a bag of dog food for $25*1.04^800699 = $10^13700.
Will I retire or break 10K?
The "Computers can play chess better than humans so they must be intelligent" argument drives me nuts.
Show me a damned computer that can *invent* a game like chess, then I'll be impressed. Should I call my calculator intelligent because it can figure the square root of 209384298332 faster than I can?
A chess program is good at following a set of chess rules (designed by people). End of story.
What were you expecting?
...is to be used to scan the net for trademark infringements. Sad, so sad.
This is one of the arguments Jaron Lanier makes in One half a manifesto. The Turing test has already been passed, the machines are as "intelligent" as humans. But this has not happened in the way usually envisioned, namely, machines achieving human-level intelligence. Rather, humans have become less intelligent. Often when we interact with a machine, we restrict our intelligence to the machine's intellectual framework. Can you tell the difference between a bank statement or credit report generated by a human and one generated by a machine? The answer, of course, is no, and this is the kind of Turing test that is occuring every day.
body yeah.
Actually I'm 25 (26 if you round up).
I've already shown that chat bots fail when asked to describe ascii art. It will be a while until they can pass this test reliably.
I discuss the accessibility implications of CAPTCHAs in this writeup on Everything 2.
Will I retire or break 10K?
Discussed above.
Would you please quit it with the Yakov knockoffs?
not too sure computers are up to par yet:
"Result of the Test: FAIL
You entered the following words:
heart
neck
kinife
The words possibly displayed in the image were:
neck
ear
ring
heart
knife
able
foot
Many minds. Long time.
You're looking for insight on Slashdot, that's where you're wrong.
Bullshit, what's all that Insightful +5 posts that were picked up by moderators who has insight....hmm...nevermind.
Muahahahahaha! Now they'll think I'm a real person. Ooops. Did I just say that?
Go through Google News and you don't have to register. Here's a direct link to the story.
AI can manage to hold a conversation with itself... maybe then ! will we have SOMETHING..
it would show the ability to stimulate as opposed to merely responding to it.
http://www.nik.com.au/alice/
-judging another only defines yourself
typos. computers should have zero typos or something is terribly wrong!
At least with Radio Shack if you pay cash, you can remain anonymous. The point of registration with the NYTimes isn't any more of a security breach - give them a free hotmail address - they have your IP address already, and in most cases this can be traced back to you, so visiting any web site dispells your illusion of anonyminity anyway. It's moot.
But it's standard sports medicine. They not only talked about it during the Biathalon event TV coverage during the last winter Olympics but showed video of a shooter hooked up to an EKG.
You could actually watch the heart stop as they pulled the trigger.
I'll have a poke around and see If I can refer you to any references, on or off the web.
KFG
My first thoughts as I read this article were of the Turing Test. I don't know if I could handle it if spammer tools became smart enough to pass a test that I myself have with on bad days.
Does this
the question of whether computers use intelligence the same way as humans use intelligence has long been determined through the 'chineese room'.
the point of John Searle's Chinese Room being is to see if 'understanding' is involved in the process of computation. if you can 'process' the symbols of the cards without understanding them (since you're using a wordbook and a programme to do it) - by putting yourself in the place of the computer, you yourself can ask yourself if you required understanding to do it:
Minds Brains and Programmes (The Original Chineese Room):
http://www.bbsonline.org/documents/a/00/00/04/8
the complementary question - 'is the human brain
a digital computer' is answered by the same author:
Is the Human Brain a Digital Computer (John Searle):
http://www.ecs.soton.ac.uk/~harnad/Papers/Py104
Summary of the Argument:
1. On the standard textbook definition, computation is defined syntactically in terms of symbol manipulation.
2. But syntax and symbols are not defined in terms of physics. Though symbol tokens are always physical tokens, "symbol" and "same symbol" are not defined in terms of physical features. Syntax, in short, is not intrinsic to physics.
3. This has the consequence that computation is not discovered in the physics, it is assigned to it. Certain physical phenomena are assigned or used or programmed or interpreted syntactically. Syntax and symbols are observer relative.
4. It follows that you could not discover that the brain or anything else was intrinsically a digital computer, although you could assign a computational interpretation to it as you could to anything else. The point is not that the claim "The brain is a digital computer" is false. Rather it does not get up to the level of falsehood. It does not have a clear sense. You will have misunderstood my account if you think that I am arguing that it is simply false that the brain is a digital computer. The question "Is the brain a digital computer?" is as ill defined as the questions "Is it an abacus?", "Is it a book?", or "Is it a set of symbols?", "Is it a set of mathematical formulae?"
5. Some physical systems facilitate the computational use much better than others. That is why we build, program, and use them. In such cases we are the homunculus in the system interpreting the physics in both syntactical and semantic terms.
6. But the causal explanations we then give do not cite causal properties different from the physics of the implementation and the intentionality of the homunculus.
7. The standard, though tacit, way out of this is to commit the homunculus fallacy. The humunculus fallacy is endemic to computational models of cognition and cannot be removed by the standard recursive decomposition arguments. They are addressed to a different question.
8. We cannot avoid the foregoing results by supposing that the brain is doing "information processing". The brain, as far as its intrinsic operations are concerned, does no information processing. It is a specific biological organ and its specific neurobiological processes cause specific forms of intentionality. In the brain, intrinsically, there are neurobiological processes and sometimes they cause consciousness. But that is the end of the story.
--
best regards,
john
Actually, pad're the computer cannot calculate exactly the "3-body" problem ... which problem is a perfectly deterministic physical system. Not all deterministic physical problems are computable. Go back to the closet and shut up, byte-boyz ... Eh ??
Sometimes the human learns chinese all by himself/herself.
;).
Sometimes just a bit of motivation is enough (not even training).
And seems kids do it better
Trouble is we may be able to model it but it doesn't mean we understand it.
At least when we are simulating/modelling weather we can start with base points and do comparisons.
Whereas with stuff like consciousness, I suspect even if the model is broken you might not be able to tell till much later.
If we really wanted intelligent entities which we didn't understand (how they work), there are always humans and other creatures.
The GM bunch may even concoct a few more.
The question raised is if human intelligence is of the same /kind/ as computer intelligence. I.e. if we build a computer big enough, will it only better *mimic* a human, or will it actually gain understanding about what it's doing? Without understanding, a human can eventually spot the entity as being a computer.
/all/ relevant material though to make a translation (if you see this, write this, etc). The guy (can) give output of the same quality of that of the real translator. But does he have any clue about what's going on? No. He only replaces a serie of chinese symbols by english (roman) symbols.
It's like the Chinese Chamber experiment.. you lock an english/chinese translator in a room, and hand him pieces of chinese. He translates them into english.
Now suppose you put a person in that room who doesn't know a thing about either chinese OR english. You give him
The same can be said about machines. They can be made to give the same output on the same input compared to a human being, more and more. But will it ever have a clue or even self-awareness so we can actually create a human being?
Or, in this case, how far do we have to go to fool others into thinking it is a human?
I thought shooters just wait for the right moment between heart beats and breath to shoot. They just make sure they are very fit and so their hearts beat slowly.
I only dimly recall reports of one guy being able to stop and more importantly RESTART his own heart. And he wasn't an olympic shooter. Plus I doubt you'd do these things standing up - guess what happens to your blood flow. Can't find links at the moment.
I'm not saying it's impossible, but I believe it's rare. Stopping one's heart voluntarily doesn't seem as useful (or safe) as being able to open your eustachian tubes voluntarily.
Then again maybe it's not rare, maybe there were people who _finally_ figured out how to stop their hearts neuro-voluntarily but unfortunately failed to restart their hearts and failed. So all we hear = death unknown causes. Not like you get much practice restarting your heart from zero y'know.
Didn't you read the article or even the story blurb?
Computers ARE telling the difference between humans and computers.
If the human and machine have a child who is also fertile, then they are the same species already.
All the ones I've used are terrible in some maJOR WAY! (not mention we can't even fix the location of the caPS LOck keY!.
Research such as this is important to enable the uman resistance to communicate securely after the machines have taken over...
...and he grinned, like a fox eating shit out of a wire brush.
Why do we need to simulate human intelligence? It is not useful. There are already enough humans to behave humanly, if you would like. A human-like intelligence will incorporate all aspects of humanity (and animals), including emotions, autonomy insticts for self-preservation etc.
Do you really think that robots would be something like Data in Star Trek? I think this is really a cliche that will never be true. We need systems that behave intelligently, but whose needs will not be as our own. Intelligence may not be separated from the underlying biological mechanisms that constitute the organism as a whole.
In the end, we shall be seeing a lot of intelligent systems that perform interesting, useful, but very specific tasks. Already some have emerged. As for human-like intelligence, recent advancements in neurobiology and computational learning theory seem to indicate that it is at least theoretically possible right now. However, as I said before, creation of a human-like intelligence is of purely academic interest and has no practical purpose. This is why we might never see it happen.
I miss my rubber keyboard.(Homepage)
you got me, I didn't read the article first. However; after reading it I still think my original idea is valid.
:P
I didn't expound on it, but the test given in the article is a single point test, not a volley of different determinations as is used in the turing test. In the turing test you have 5 minutes and a simple text display/intermediary, in this you have a single visual problem to solve.
What I was proposing was that you take the original confines of the turing test and replace the human interigator with a computer, where the communication mechanism is still a teletype or intermediary.
There is one thing the test in the article overlooks. As of now I believe that there are a number of AI systems that have been put to the turing test that behave similar to young children, including those that can't read. In the original turing test this isn't an issue. Also the articles method is not actual AI, but a simple "puzzle", I think it would be more interesting to see and actual AI agent that could tell the difference, or an AI agent that developed this "puzzle"
"Not knowing when the dawn will come, I open every door." - Emily Dickinson
John Searle is an idiot, and nothing has "long been determined" by the Chinese (yes, two e's. Not 3.) Room exercise. His argument hinges on the fact that the MAN in the room doesn't understand Chinese, which is an inappropriate analogy. The "man + rulebook + room" DOES understand Chinese. The man himself doesn't need to. It's like asking if your amygdala understand English.
I'm sorry that you can't think for yourself, and need Searle to do it for you. Maybe you're one of the few to whom his results apply. As you reread your post, "ask yourself if you required understanding to do it".
Is there some kind of time limit? Apparently I'm not supposed to wait for the image to load. I got this result 5 times in a row. Maybe it's the ol' Slashdot effect fucking with the latency.
Too bad you didn't provide the link for the crackerjack box from which you read that. The counter argument you provide is flawed, unfortunately.
..machines must either make the same leap in complexity such that their actual thought processes can no longer be mapped directly to the underlying formal system..
The thing about this is, the difference between one and the other doesn't necessarily lie in any physical complexity. The idea is that a system of sufficient complexity should develop higher level processes that aren't directly mappable to the structure all by itself. Take the ant hill, for instance. Hofstadter explained it quite sufficiently, so I won't.
So you see, computers don't need to make any "leap in complexity". You're invoking images of some kind of mystical metamorphosis. When and if a computer is made of sufficiently complexity and linkage, these "actual thought processes" (as opposed to some kind of false thought process?) will manifest themselves quite apart from the designer's intention.
And those manifestations will probably be considered bugs. Mandelbugs. So could it be that computers are already beginning to reach sentience, and that the ones that are furthest along are always "fixed", thus resetting the process? Such a thing, left to itself, would not just accrue sentience. Computers are at a disadvantage because they cannot gather input of their own free will. They can only gather input when they're being used, unless they're specifically set to do it by automation. Nobody wants to waste the power of the ASCI White by reconfiguring it with simple data rearrangement rules and setting it to parse the internet by itself, just on the off-chance it will become self-aware. After all, what is sentience except the ability to respond to input in a meaningful way by referring back to previous input that is somehow similar? It is in how "similarity" is judged that the fuzzy issues come into play.
A child who has never been taught anything in its life, including language, will not be able to understand that "It thinks, therefore it is." Is it, then, any less sentient? Even if it did understand, how would an external interrogater learn that? The child doesn't understand language. Even if you teach it to push bits according to an external rule system, it won't necessarily be able to use this knowledge to convey it's own thoughts about sentience. Chinese room scenario again. The child is sentient, but it doesn't have to understand what it's doing with the bits to do it right. So it wouldn't be able to use this method of output as a way to declare its sentience. The child+rulebook understands, but the child does not. So we'd have to wait until the system of the child+rulebook developed sentience of its own, apart from that of the child. And would THAT sentience understand that the child was sentient? Not necessarily. Not to mention that the relationship between the child and the rulebook is so simple, that system would never devlop sentience.
Or perhaps you subscribe to the "Breath of Life" school of thought, that claims some magical "extra" is necessary for sentience. Puh-lease. I'm going to shoot the next person that says "only God can create life".
I think that that poll was indeed representative of the best graduate school in CS, after all, MIT won while coming in one day after CMU, thus proving that their bot was better and, by extension, that the product of their CS courses were better CS engineers than those at CMU.
And let's not talk of all these other CS schools who weren't even able to put a bot together to compete.
"The obvious mathematical breakthrough would be development of an easy way to factor large prime numbers." Bill Gates,
IMO the biggest problem would be false positives, when the program takes (stupid?) humans for AI.
But what if there was no difference between a good AI and a stupid human? I guess that would be some level of sucess for AI's. The ultimate goal being to create an AI that humans, and AI's, can't detect. A lot of current AI systems that do the Turing test end up sounding like little children. Is this a bad thing, or just the first step?
"Not knowing when the dawn will come, I open every door." - Emily Dickinson
please moderate it accordingly
Actually actually, the 3-body isn't a deterministic system in a physical sense any more than it is in a computational sense. The inherent problem with the 3-body problem is that in general, solutions are unstable because they depend greatly on specific initial conditions, which can only be known so well.