The keys to the digital kingdom are credentials. In no industry is this more true (and ultimately more damaging) than financial services. The sophistication of the attacks used to gather those credentials and thwart the increasingly complex authentication process that guards financial transactions is sometimes staggering. That's because they require not just stealth but coordination across multiple touch points in the transaction process.
MITB is not a new technique. It was first identified as a potential path to financial theft back in 2005 when Augusto Paes de Barros presented it as part of his "The future of backdoors - worst of all worlds". MITB didn't receive its "official" title until it was so named in 2007 by Philipp Gühring. In 2008, Trojans with MITB capabilities began to surface: Zeus. Torpig. Clampi. Citadel.
Most financial targeting Trojan malware are able to capture a wide variety of behavior and data as well as enabling remote control via VNC and RDP. They range from keylogging and grabbing and more recently have begun taking advantage of the mobile explosion in order to bypass multi-factor authentication requirements that leverage SMS to deliver OTP or TAN to legitimate users.
These Trojans accomplish these feats using MITB to inject scripts into legitimate banking web applications to extract and deliver credentials to dropzones. These scripts are dangerous not only because of the amount of data they can collect; that's true of just about any Trojan that inserts itself into the user environment. These scripts are dangerous because they become part of the application logic. MITB essentially changes the client side of a web application, giving it potentially new and dangerous capabilities such as modifying the details of transactions in real time. You might think you'd notice that, but because it's in the browser and modifying the business logic it can hide those changes from you, at least for a few days.
Financial institutions attempting to put a stop to these fraudulent activities often implement two-factor authentication. Logging in with simple credentials is not enough; a second password or code is delivered via a secondary channel such as via SMS in order to authenticate the user.
But even this is not always enough. As our F5 SOC research team recently showed, Neverquest is able to abuse user's trust in their banking institutions against them. In addition to the typical MITB script injection designed to steal credentials, Neverquest attempts to coerce users into also installing an application on their mobile device, designed to capture and deliver secondary authentication codes and passwords as well. Successfully doing so means attackers can execute automated transactions against the users financial accounts.
Given that 73% of users are unable to distinguish between real and fake popup messages (Sharek, et al 2008, "Failure to Recognize Fake Internet Popup Warning Messages") the potential for Neverquest and similar Trojans to succeed in such efforts is more likely than not, particularly when those popups are presented by a trusted site such as a financial institution.
The key to detecting these script-injecting, app modifying monsters is to understand the state of the web application page at the time it's delivered - before the Trojan has a chance to modify it - as well as monitoring for duplicate communication initiated from the web page. These are both methods used by web anti-fraud solutions to detect infected clients. A small protective script is included in each and every page that can detect attempts to modify the logic as well as notice duplicate communications and can notify the user immediately.