How can I tell you're not a robot?

In the virtual world of the Internet, identities often get blurred. Senior citizens easily pass themselves for teenagers, and timid geeks, normally hiding behind their glasses, can take on the roles of seductive young women. If you are not careful enough when chatting with strangers, you might soon find out that the charming young woman you have shared so many good talks with is, in fact, none other than your annoying neighbor.
The development of the Internet quickly revealed another problem concerning identification, one that can almost be considered a philosophical issue. Not only can people virtually take on other identities on the Web than they possess in real life, computers themselves can impersonate people. Today, this most certainly is not a thing of science fiction as these identity swaps occur on a regular basis. Robots or, to be precise, programs portraying other people online, sign in to all kinds of Web forums where they post advertisements for their masters or creators. The smarter ones even know how to create their own free e-mail accounts which they use to flood other accounts with advertisements every day.

Exterminator wanted
Robots roaming the Internet and portraying other people have become a serious problem around 2000. Administrators of larger websites realized they needed to invent an effective method of distinguishing real people from robots as soon as possible. The task seems simple enough at first, but it proves to be a pretty tough nut to crack as soon as you take it on. Should websites employ people who would exchange a few words with every single visitor who wants to sign in? This would certainly prove to be a waste of time and money. A more effective solution was needed.
The job description of a person detecting robots who surf the Web creating problems for website owners would not be anything like that of the legendary Blade Runner where officer Rick Deckard (Harrison Ford) chases runaway replicants or robots who look and act just like people. However, this comparison is not completely out of place. The movie implies that officer Deckard himself is a replicant. The development of programs that can differentiate between robots and human beings on the Internet was closely followed by searching for a method of exterminating these troublesome artificial creatures that wreak havoc on the Web.

The Turing test
Soon after the Second World War, Alan Turing, an English mathematician who contributed his share to constructing the first computers, contemplated the possibility of computer thought. He quickly found out this was a much too abstract question, so he tried to transform it into something more tangible and simple. He started to develop a practical method of differentiating between artificial and human intelligence. How to determine whether a machine is capable of thinking? One of the most obvious solutions is talking to this machine. If it is impossible to tell the difference between its responses and those given by a human being, then there simply is no more difference between artificial and human intelligence. The Turing test, as it was later named, is, simply put, a method of determining whether someone is a man or a machine. Both can be posed questions to which they have to answer. At the same time, we have no notion about who is in which room. If after a longer interview we are unable to determine which room conceals the living person, the other entity has successfully passed the test and, according to the Turing criterion, cannot be set apart from a real human being. The Turing test is thus a way to determine, by exchanging a series of questions and answers with two subjects via our keyboard, if we are conversing with a man or a machine. The machine passes the test when the assessor cannot tell whether he is talking to a man or a machine.

What robots can’t do?
Disturbing attacks by malicious robots have forced the administrators of larger websites to seek help from experts, specialized in artificial intelligence. They quickly concluded that the best solution for "exterminating" such robots would be a series of brief tasks that real people would find easy to solve in a matter of moments, while computers or robots would need more time to succeed, if they could at all. Computers are much faster in solving simple calculations than humans, so mathematical tasks were not an option.
It soon became clear that the tasks should involve problems which pose the greatest difficulties to artificial intelligence researchers. One of these was the visual recognition of images and symbols. This is where people have, during millions of years of evolution, developed extremely efficient means of recognizing images and symbols even when they are distorted.
Modern computers are becoming quite good at recognizing an image of a text. They can determine which letter is portrayed just by scanning the form presented by the image. This is how they are able to convert the copy of a printed text back into a file which can then be edited and changed anew. However, such recognition can only be successful if the image of the text is sufficiently clear and legible. If the letters are smudged, these programs usually turn out to be helpless. They can only see an indistinct blotch and not the text, while people can still read damaged or smudged writing.

The Automated Turing test
This is how engineers came up with the idea to create a robot, capable of differentiating between artificial and human intelligence. While people have no difficulties reading smudged letters and numbers, that is something artificial intelligence of today can’t cope with yet. The researchers of the Carnegie Melon University have developed an automated Turing test that needs no human guidance. They named it CAPTCHA, which stands for Completely Automated Public Turing Test to Tell Computers and Humans Apart.
In simpler terms, CAPTCHA is a program that can compose and evaluate a test that it could not actually solve by itself. It is a program for exterminating programs. It gives us a simple visual task that usually requires us to copy a series of smudged letters and numbers, shown on the screen. According to our answer, the program can tell whether we are real humans or mere robots.
Every Internet user has probably encountered such a program at some point. We have to pass its test in order to register to an Internet forum, to create a new e-mail account or to file our income tax return. Of course, as years go by, artificial intelligence is becoming increasingly advanced and programs for exterminating programs have to be improved all the time.

Which is the best school for computer science?
There is an entertaining anecdote about how robots can participate in an Internet survey from times when programs for exterminating programs did not yet exist. In November 1999, the website slashdot.com started a poll to find out which was the best school for studying computer science which is quite a delicate question, especially when posted on the Internet.
To make voting as fair as possible, each computer was given one vote. The program that registered the votes also registered Internet addresses (IP addresses) of individual computers that had already cast their vote, allowing each individual to vote only once. However, students from the Carnegie Mellon University found a way to bypass this protection and designed a program which cast thousands of votes for their university. By the next day, MIT students wrote their own voting program which gathered votes for their side. This is how the battle for votes between the two robots/computer programs began. In the end, MIT snatched its victory by 21.156 votes, Carnegie Mellon followed with 21.032 votes, and other schools finished with less than 1000 votes each.
When it comes to deciding on the best school for computer science it might be completely appropriate to grant computers an equal right to vote. In the case of other surveys, however, one would probably be much more comfortable knowing that each registered vote represents a human being.