Bot ‘breaks’ Captcha, making the most annoying thing on the internet pointless
An artificial intelligence system has managed to solve 'a wide variety' of challenges with 'very little training data'
Your support helps us to tell the story
From reproductive rights to climate change to Big Tech, The Independent is on the ground when the story is developing. Whether it's investigating the financials of Elon Musk's pro-Trump PAC or producing our latest documentary, 'The A Word', which shines a light on the American women fighting for reproductive rights, we know how important it is to parse out the facts from the messaging.
At such a critical moment in US history, we need reporters on the ground. Your donation allows us to keep sending journalists to speak to both sides of the story.
The Independent is trusted by Americans across the entire political spectrum. And unlike many other quality news outlets, we choose not to lock Americans out of our reporting and analysis with paywalls. We believe quality journalism should be available to everyone, paid for by those who can afford it.
Your support makes all the difference.Researchers have created an artificial intelligence system that can solve Captcha challenges, rendering them “broken” and “ineffective”.
The security check is designed to block potentially harmful bots from websites, and does so by presenting puzzles that are supposed to be easy for people to solve, but very difficult for computers.
They’ve been around since the late 1990s and have long been considered extremely annoying, but experts believe it could soon be time for them to be replaced.
The researchers, from a company called Vicarious, managed to build an AI system – called the Recursive Cortical Network (RCN) – that approaches Captcha challenges in much the same way that a human would.
“With one model, we achieve an accuracy rate of 66.6% on reCAPTCHAs, 64.4% on BotDetect, 57.4% on Yahoo, and 57.1% on PayPal, all significantly above the 1% rate at which CAPTCHAs are considered ineffective,” they wrote in a blog post.
“When we optimize a single model for a specific style, we can achieve up to 90% accuracy.”
The researchers say it managed to break “a wide variety of text-based Captchas” with “very little training data”.
Instead of trying to familiarise itself with every possible type of Captcha it might come across, the researchers wanted the system to “learn and generalize from a few examples”, as humans can.
“Although specific Captchas have been broken before using style-specific segmentation heuristics, those attacks could be foiled easily by minor alterations to Captchas,” the researchers wrote in a paper that has been published in the journal Science.
“RCN breaks the segmentation defense in a fundamental way and with very little training data, which suggests that websites should move to more robust mechanisms for blocking bots.”
Join our commenting forum
Join thought-provoking conversations, follow other Independent readers and see their replies
Comments