We’ve all struggled with those often unreadable security Captcha codes from time to time, but if some new research out of Stanford is any clue, machines might be better at it than us very soon. By using machine vision algorithms, the Stanford team was able to defeat 66% of Visa’s Captchas, 70% of those used by Blizzard, and about a quarter of Wikipedia’s Captchas. This may spell trouble for a multitude of other sites, as well.
While some of these numbers might not sound like a huge problem, the team suggests that any hit rate over 1% means Captcha is too broken to continue in use. Since the cost associated with having a computer try again and again to solve the Captcha is so low, those with malicious intent could blow through this security measure with no problem. Most companies use various layers of security, but Captchas are relied upon heavily for weeding out bots.
The Stanford team had no luck whatsoever with Google’s ReCaptcha system, which is used widely. This indicates that the visual filtering and de-blurring techniques aren’t entirely there, but its success with other systems could mean it’s just a matter of time.