Google recently announced their “no CAPTCHA reCAPTCHA” method that simplifies the end-user experience while offering an intelligent threat assessment behind the scenes to detect potential robots. All you will typically need to do is click a checkbox that tells their CAPTCHA that you’re not a robot. Then a brief loading animation occurs while it runs the assessment algorithm. If the system is not sure you’re human, it will then prompt you with an additional CAPTCHA check that uses the traditional distorted image for an extra security check. However, if you’re on a mobile browser, you may encounter a different method for the secondary check. This new method that Google is experimenting with will display an image, such as a cat, and will ask you to select all other images that match the subject.
So I definitely see the positive side here since humans should be able to proceed more quickly through a signup form or other checkpoint that requires CAPTCHA. People are impatient when using the internet, so every effort to speed things up is helpful in improving their experience. However, the other point Google makes is that robots are being programmed with enough intelligence to decipher these distorted text images with a 99.8% accuracy. So given that they’ve added an extra step with the click a checkbox, the next step isn’t any different than regular CAPTCHA unless you get the occasional experimental CAPTCHA that uses an image matching test. I’m not aware of the programming that goes into these robots, but it seems like it would be easy enough to tackle a checkbox and then interpret the distorted text image as usual.
What do you think?