SNEAKY SPAMMERS have managed to crack Gmail's protective CAPTCHA, allowing bots to automatically sign up for new accounts, meaning users face even more spam attacks.

Websense Security Labs reckons that cracking the CAPTCHA (Completely Automated Public Turing test to tell Computers and Humans Apart ) was no easy feat and that it required a minimum of two bots, unlike the single bot that was required to recently crack the Windows Live Mail CAPTCHA.

Websense also reckons that the same group of hackers was behind both security breaches. The double bot breach works by having the two bots work together, checking the other's work and ensuring that if one bot times out, the other can continue cracking the code.

The system only seems to have worked once every five
CAPTCHAs, in which users are made to decipher a series of eye squintingly unfathomable letters and numbers embedded inside an image, are supposedly even more befuddling to computers than to humans themselves. But, hackers seem to be rapidly catching up with CAPTCHA code.

Lots of effort is now being put into making stronger, harder to hack CAPTCHAs including audio-based ones (requiring the user to type the word heard back into a text box), and complex image ones which make the user choose a set of similar images (which are nearly always kittens or puppies) before being allowed to go any further. These more muscular CAPTCHAs discriminate against hearing and sight impaired people, but are deemed
necessary to prevent the codes from repeatedly being hacked. µ

L'Inq
Websense