Sadly, one of the First World Problems are CAPTCHAs. They are essentially simplified Turin tests aimed at preventing Web resource abuse. However, most of the abuse is felt by the user with good intentions as he stumbles upon this:
Furthermore, most types of CAPTCHA schemes have already been overcome by successful types of attacks, rendering their existence at least debatable.
However, researchers at the University of Alabama at Birmingham ( UAB) have developed a system that could forego all the annoyances while offering increased security. They focused on the so-called dynamic cognitive games ( DCG) that require the user to perform a gamelike cognitive task by interacting with dynamic images. One example required the user to identify a boat from a set of moving objects and then drag-and-drop it in a designated “dock” location.
This task, however simple for us humans, puts a great strain on a robotic mind. It also provides human users with a much more pleasant experience. The team of researchers, led by Prof. Nitesh Saxena set out to test the effectiveness of DCG CAPTCHAs. They first created a few prototype games and then developed a fully automated attack framework in order to break them. As expected, the robots did not perform well.
Another disadvantage of present-day CAPTCHAs is that they fall victim to relay attacks. These are ordered by criminals paying as little as one penny/cracked CAPTCHA to people willing to crack them. Of course, this problem would occur even with DCG protection but researchers say the detection of said relay attacks is more reliable using DCGs.
Currently, the scientists are working toward redesigning DCGs with the help of companies like Are You a Human.
I, for one can’t wait to permeate eQ7fltW.