Latest research shows that robots are faster and more accurate than humans at cracking 'anti-bot' verification codes

News from this site on August 13, according to a new study that referenced the security systems of more than 200 popular websites, bots are better than humans at passing bot tests. However, the research has not yet been peer-reviewed and is currently only published on arXiv.

Bots pose a major threat to the Internet because they can disguise themselves as legitimate users and perform operations that endanger the harmonious environment of the Internet, such as crawling content, creating accounts, posting false comments or comments, and consuming scarce resources.

Scientists, including those at the University of California, Irvine, point out that "if unregulated, bots can perform these malicious operations on a massive scale."

In the more than two decades since the birth of CAPTCHA, most websites have deployed it as a security check strategy, through some puzzle that "should be very easy for humans but very difficult for computers" to block potentially harmful bots.

This site noticed that early verification codes required users to identify and retell text from distorted images, but with the advancement of computer vision and machine learning technology, it only took a short time for the robot to successfully complete the task with nearly Recognize text within it with perfect accuracy, faster and more accurately.

Subsequently, humans began to engage in an arms race with robots in the field of verification codes. Because of this, CAPTCHAs are starting to become more and more cumbersome and annoying, even for humans and bots to solve.

In this study, scientists tested 200 of the most popular websites and found that 120 of them still use traditional CAPTCHA mechanisms. .

They conducted 10 CAPTCHA tests on these sites to assess their difficulty, with the help of 1,000 participants from different backgrounds (location, age, gender and education level).

Many robots described in scientific journals can outperform humans in speed and accuracy, researchers have found.

For example, some CAPTCHA tests can take a human participant 9 to 15 seconds to solve, with an accuracy of about 50% to 84%, while a bot can solve it in less than a second It is cracked within and the accuracy is almost perfect.

"The accuracy of the robots ranged from 85-100%, with most exceeding 96%. This significantly exceeds the range of human accuracy we observed (50-85%)," scientists in the study wrote.

They also found that in all cases the time it took the robots to solve the problems was "significantly lower" or nearly the same as humans.

Advertising statement: The external jump links (including but not limited to hyperlinks, QR codes, passwords, etc.) contained in the article are used to convey more information and save selection time. The results are for reference only. All articles on the site contain this statement.

The above is the detailed content of Latest research shows that robots are faster and more accurate than humans at cracking 'anti-bot' verification codes. For more information, please follow other related articles on the PHP Chinese website!