Security Experts:

Google Uses New Approach to Simplify CAPTCHA Solving

A new type of CAPTCHA system introduced by Google is efficient in preventing spam and abuse, and makes it easy for users to verify that they are human, the company announced on Wednesday.

CAPTCHA systems, including Google's reCAPTCHA, have relied on distorted text to filter out robots and abusive scripts. However, artificial intelligence technology is highly advanced and recent research by the search engine company showed that even the most distorted text can be deciphered with a 99.8% accuracy.

In an effort to make the verification process easier for humans and more difficult for bots, Google launched the "No CAPTCHA reCAPTCHA," which enables users to verify that they're human without actually having to solve a CAPTCHA.

Users who encounter Google's CAPTCHA will be able to demonstrate, in most cases, that they're human simply by checking a box. An additional CAPTCHA will only appear if the company's risk analysis engine isn't totally convinced that the user is not a bot.

"While the new reCAPTCHA API may sound simple, there is a high degree of sophistication behind that modest checkbox," Vinay Shet, product manager of Google's reCAPTCHA, explained in a blog post.

"Last year we developed an Advanced Risk Analysis backend for reCAPTCHA that actively considers a user’s entire engagement with the CAPTCHA—before, during, and after—to determine whether that user is a human. This enables us to rely less on typing distorted text and, in turn, offer a better experience for users," Shet said. "The new API is the next step in this steady evolution."

It's often difficult for users to solve CAPTCHAs on mobile devices, which is why Google says it's experimenting with some new mobile-friendly challenges. In one example provided by the company, users are presented with a clue and they're asked to select some images that correspond with it.

Snapchat, Humble Bundle and WordPress are already using the new API to make things easier for their customers. According to Google, in the last week, 60% of visitors and 80% of visitors were saved the trouble of trying to decipher distorted text.

In a SecurityWeek column published last year, Tal Be’ery, VP of Research at Aorato, made some recommendations for improving the system and Google appears to be heading in the right direction.

"Application owners should augment their current CAPTCHA based anti-automation protections with some non-customer-facing anti-automation protections to reduce negative user experience and the dangers of interactive attacks. CAPTCHA should be used scarcely and as the final anti-automation measure, and not as the first and only anti automation measure," Be’ery said.

view counter
Eduard Kovacs (@EduardKovacs) is a contributing editor at SecurityWeek. He worked as a high school IT teacher for two years before starting a career in journalism as Softpedia’s security news reporter. Eduard holds a bachelor’s degree in industrial informatics and a master’s degree in computer techniques applied in electrical engineering.