A technology start-up claims that it has developed a software that works like a human brain and has the ability to read CAPTCHAs.
CAPTCHA (Completely Automated Public Turing test to tell Computers and Humans Apart) is a program developed in 2000 by students from the Carnegie Mellon University for Yahoo. It can generate and score tests that humans can pass but computers can't. It usually gives distorted texts that humans can read but computer programs can't.
A San Francisco-based start-up Vicarious was able to develop a software that can read CAPTCHAs with a success rate between 90 and 97 percent. They clarified though that it is not for sale and they have no intent of using it for any illegal purpose. They developed the software to attract investors interested on artificial intelligence.
"We wanted to show we could take the first step toward a machine that works like a human brain, and that we are the best place in the world to do artificial intelligence research," co-founder D. Scott Phoenix told Reuters.
"We [train] the system by showing it images of letters," says Dileep George, Vicarious' other co-founder, to Mashable. "It needs just a few examples of letters to learn about them. Previous work would require in the order 10,000 examples of a letter even to understand minor variations."
Vicarious had not published any paper describing its methodology which made security experts skeptic about the claim. One of the CAPTCHA developers, computer scientist Luis von Ahn, is one of those doubting the start-up.
"CAPTCHAs have been around since 2000, and since 2003 there have been stories every six months claiming that computers can break them," Luis von Ahn told Reuters. "Even if it happens with letters, CAPTCHAs will use something else, like pictures" that only humans can identify against a distorting background.
Vicarious hopes to receive more support and generate more funding so they can expand artificial intelligence to robotics, medical image analysis, image and video search, and others applications.
