Google Finally Puts Captchas to Good Use

Illustration for article titled Google Finally Puts Captchas to Good Use

Captchas are the Internet's speed bump. They perform little function beyond slowing users down—not bots, mind you, just people. But now Google's found a way to employ the oddly-fonted phrases for a different kind of verification.


ReCAPTCHAs are the second generation of Completely Automated Public Turing Test To Tell Computers and Humans Apart, designed at Carnegie Mellon University and purchased by Google in 2009. In addition to blocking bots, these challenge phrases also assisted Google's text digitization efforts. ReCAPTCHAs traditionally only display text-based challenges yet, as the image above shows, have also begun adding image-based challenges too. But why do the images all look like cropped Street View shots? Because they are.


According to Google,

We're currently running an experiment in which characters from Street View images are appearing in CAPTCHAs. We often extract data such as street names and traffic signs from Street View imagery to improve Google Maps with useful information like business addresses and locations. Based on the data and results of these reCaptcha tests, we'll determine if using imagery might also be an effective way to further refine our tools for fighting machine and bot-related abuse online.

So basically, Google's crowd-sourcing data verification services that would be more expensive and labor-intensive if done by computers and imaging algorithms. Not only does it provide a new means of fighting spam and bots, it will also make Google Maps more accurate—that's a win-win for Google and users. [TechCrunch - Image: Dirtbag / Blackhatworld]

Share This Story

Get our newsletter


RuBBa_cHiKiN: North Gizmodo is Best Gizmodo

Google's Captchas typically include one actual word problem either preceded or followed by a scanned image of a word from a book. The only one that is actually checked is the warped random text. The other one is not checked, but whatever word is typed is used to identify that image's contents via crowdsourcing. Therefore, you can type whatever the hell you want for the scanned image so long as you get the actual puzzle correct.

Therefore, I have always typed "bacon" for the scanned image.