The Secret Side of Captcha

The Secret Side of Captcha

Though Captcha has become a common sight for most internet users, what some may not realize is its uses have included more than sorting humans from spambots. Captcha was invented in 2000 by Luis von Ahn and his professor, Manuel Blum. It started out as a single distorted word, to be typed into a box to differentiate humans from computers, but its next form, which von Ahn released in 2006, featured two words. One was simply a distorted word, but the other was a word from a digitization project, from a book or newspaper that computers couldn’t translate into text. By 2008, ReCaptcha had transcribed over a billion words, and within a couple of years, it had digitized the entire archives of the New York Times, as well as a large portion of Google books. As computers got better at reading the text (by 2014 they were better at it than human users), Captcha switched to images on Google maps, helping to identify and tag street signs. In 2014, it also began asking users to choose between images; this data was then collected to teach computers to recognize objects. As Captcha grew more advanced, however, its double use disappeared in favour of convenience. Thanks to Captcha’s ability to sort computers from humans by looking at mouse movements and cookies, the program is now able to spot bots invisibly without making a human click or type anything, unless the program is unsure.

Source: The Secret Side of Captcha

Leave a Comment