Risk
2/19/2009
12:46 PM
Keith Ferrell
Keith Ferrell
Commentary
Connect Directly
RSS
E-Mail
50%
50%

CAPTCHA Cnondrum: Automated Attacks Trump Human-Entry Defenses

Automated attacks aimed at bypassing CAPTCHA -- those squiggly characters you have to enter to access some blogs and e-mail -- are getting better and faster at overcoming anti-spam defenses. In other words, the machines are beating us at what was supposed to be our game.

Automated attacks aimed at bypassing CAPTCHA -- those squiggly characters you have to enter to access some blogs and e-mail -- are getting better and faster at overcoming anti-spam defenses. In other words, the machines are beating us at what was supposed to be our game.CAPTCHA -- Completely Automated Public Turing Test to Tell Computers and Humans Apart -- approach to granting access to humans only (theoretically only humans could read and enter the distorted characters) has run into increasing bot problems.

Bots, it turns out, and more specifically bot-makers, are pretty good at putting together coordinated attacks that decipher the CAPTCHA characters and -- presto! -- get access to the targeted service, create accounts on that service (Hotmail, for instance) and use that address to dispatch waves of spam.

The current bot approach is troubling not just for its ability to overcome CAPTCHA defenses, but also because communications between the bot and a central server are encrypted, raising the attack sophistication stakes another level. The server does the deciphering, the bot receives and then enters the deciphered characters.

CAPTCHA bypasses are nothing new, of course, nor is there a lot CAPTCHA makers can do, other than continue to refine and reinvent their defenses. (Don't know about you, but for my tastes the trend toward making CAPTCHA images more and more vague, fuzzy and distorted has gone about as far as it can go -- I've seen some that machines might be able to figure out, but that were beyond my ability to decipher and enter.)

Identity verification -- simple verification that you're a real person -- keeps hitting these walls, as recent news of a facial recognition technology hack shows.

The point of all this for small and midsize businesses -- unless you're in the CAPTCHA business -- is simple: the bad guys are smart, automated, sophisticated, but for botnets to work they must first get access to vulnerable machines. Sealing your defenses against compromise may not solve the CAPTCHA conundrum -- not sure anything can -- but it ensures that you and your company's computers won't, at lest, be part of the problem.

Comment  | 
Print  | 
More Insights
Register for Dark Reading Newsletters
White Papers
Cartoon
Current Issue
Dark Reading Must Reads - September 25, 2014
Dark Reading's new Must Reads is a compendium of our best recent coverage of identity and access management. Learn about access control in the age of HTML5, how to improve authentication, why Active Directory is dead, and more.
Flash Poll
Video
Slideshows
Twitter Feed
Dark Reading - Bug Report
Bug Report
Enterprise Vulnerabilities
From DHS/US-CERT's National Vulnerability Database
CVE-2003-1598
Published: 2014-10-01
SQL injection vulnerability in log.header.php in WordPress 0.7 and earlier allows remote attackers to execute arbitrary SQL commands via the posts variable.

CVE-2011-4624
Published: 2014-10-01
Cross-site scripting (XSS) vulnerability in facebook.php in the GRAND FlAGallery plugin (flash-album-gallery) before 1.57 for WordPress allows remote attackers to inject arbitrary web script or HTML via the i parameter.

CVE-2012-0811
Published: 2014-10-01
Multiple SQL injection vulnerabilities in Postfix Admin (aka postfixadmin) before 2.3.5 allow remote authenticated users to execute arbitrary SQL commands via (1) the pw parameter to the pacrypt function, when mysql_encrypt is configured, or (2) unspecified vectors that are used in backup files gene...

CVE-2012-5485
Published: 2014-09-30
registerConfiglet.py in Plone before 4.2.3 and 4.3 before beta 1 allows remote attackers to execute Python code via unspecified vectors, related to the admin interface.

CVE-2012-5486
Published: 2014-09-30
ZPublisher.HTTPRequest._scrubHeader in Zope 2 before 2.13.19, as used in Plone before 4.3 beta 1, allows remote attackers to inject arbitrary HTTP headers via a linefeed (LF) character.

Best of the Web
Dark Reading Radio
Archived Dark Reading Radio
Chris Hadnagy, who hosts the annual Social Engineering Capture the Flag Contest at DEF CON, will discuss the latest trends attackers are using.