March 11, 2008
You may have heard that spammers apparently managed to circumvence Google's Gmail Captcha (in oder to automatically create spam accounts), but did you ever wonder what the term "CAPTCHA" means after all?
CAPTCHA is an acronym for " Completely Automated Public Turing test to tell Computers and Humans Apart". The term was created in 2000 by Luis von Ahn, Manuel Blum, Nicholas J. Hopper and John Langford and has been trademarked by Carnegie Mellon University.
So a CAPTCHA is a type of challenge-response test used in computing to determine whether the user is human or some kind of an automated computer algorithm (a so-called robot). The server asks a question that only a human being should be able to answer - a common type of CAPTCHA requires that the user type the letters of a distorted image.
A CAPTCHA is in fact reverse Turing test, because it is administered by a machine and targeted to a human, in contrast to the standard Turing test that is typically administered by a human and targeted to a machine.
The so-called "Turing test" is a proposal for a test of a machine's capability to demonstrate intelligence. Alan Turing first described his test in the 1950 paper "Computing Machinery and Intelligence," it proceeds as follows: a human judge engages in a natural language conversation with one human and one machine, each of which try to appear human; if the judge cannot reliably tell which is which, then the machine is said to pass the test (the machine is then considered to be "intelligent").
As todays computers and algorithms cannot be considered to be "intelligent" when compared to a human being, they will generally fail a Turing Test.
Of course the latest events don't mean that spammers managed to create a computer system with human-like capabilities. BUT they managed to refine their algorithms to better recognize the characters in a CAPTCHA (think of a more sophisticated OCR software...), even if they have been distorted.
One of the major problems with the latest progresses is the fact that many humans already have difficulties with the currently used CAPTCHAs which are often already heavily distored and hard to read. Distorting them even more may make it impossible for many humans to read them, while advanced algorithms may still be able to read them.