Recently, a large U.S. bank client of IBM received a notification that malware was identified on one of its client’s devices. The malware was identified using a server-based malware-detection tool that identifies the presence of malware on all devices that can initiate an online banking session. The bank discovered that the user in question had not logged in to his or her bank account around the time the malware was identified, and therefore, it did not understand how malware could have been detected on the user’s device or how credential theft could have taken place.

In this case, the malware on the user’s device captured the user’s credentials at login and immediately communicated them to the fraudster’s command-and-control (C&C) center. Interestingly, the malware requested the user’s one-time password (OTP) at login, even though the user logged in from his or her regular device.

At the same time, the malware blocked the user’s credentials from being submitted to the bank. Instead, it injected a page notifying the user that the bank’s website was temporarily down (see below). Because the login request never reached the online banking server, the bank had no record of the legitimate user attempting to log in. Why would a fraudster go to such lengths rather than simply using the compromised credentials to initiate a new user session?

Injected malware message to the online banking website

Bank Risk Engines and Credentials Theft

The short answer is to evade detection by bank risk engines. Banks use these risk-based analytic tools to detect a variety of anomalous conditions that could be indicative of fraud. These risk engines are often used to identify credentials theft by looking for multiple devices simultaneously logged in to a single account, as well as successive user logins from locations that are geographically too far apart for an account owner to possibly travel within the given time frame.

When either of these conditions is met, the bank can quickly identify that fraud is being attempted and take appropriate actions. However, because fraudsters tend to be persistent and innovative, they have developed new approaches to circumvent these detection techniques.

The Risk Engine View

he following is an abbreviated representation of the bank’s log file in the example provided above:

Based on the log file, six days after accessing the account, the user logged in on an unrecognized device from a new location. Users commonly change devices and frequently travel, so this situation was flagged by the bank’s real-time risk engine for secondary authentication. The user successfully entered an OTP and was allowed to log in. However, things are not always as they appear.

The Trusteer Pinpoint View

Another view of the sequence of events can be found using an abbreviated representation of the log file generated by IBM Security Pinpoint Criminal Detection:

In this sequence, we are able to see a new, previously hidden event on line two. Because the user’s login credentials were blocked from being submitted, the bank’s online application and risk engine did not see the action indicated on line three. However, Trusteer Pinpoint Criminal Detection identified this action, which occurred four minutes before the action on line three.

Line two shows that the user attempted to log in, but the credentials never reached the bank’s server. Blocked credentials followed by a login attempt from a different machine in a distant location are clear indicators of real-time credential theft. Because the credential transmission was blocked, the bank’s risk engine only saw one new login attempt: the fraudulent one (line three).

By doing this, the criminals greatly increase the likelihood of avoiding detection and successfully committing fraud. Criminals often use a session-blocking Man in the Browser (MitB) technique to access commercial accounts that require an OTP for login. Using available malware, such as Zeus or SpyEye, cyber criminals can capture the complete set of login credentials, including OTPs, immediately log in to a compromised account before the OTP expires and block the legitimate user login attempt from reaching the bank.

More from Fraud Protection

New DOJ Team Focuses on Ransomware and Cryptocurrency Crime

While no security officer would rely on this alone, it’s good to know the U.S. Department of Justice is increasing efforts to fight cyber crime. According to a recent address in Munich by Deputy Attorney General Lisa Monaco, new efforts will focus on ransomware and cryptocurrency incidents. This makes sense since the X-Force Threat Intelligence Index 2022 named ransomware as the top attack type in 2021. What exactly is the DOJ doing to improve policing of cryptocurrency and other cyber…

What Are the Biggest Phishing Trends Today?

According to the 2022 X-Force Threat Intelligence Index, phishing was the most common way that cyber criminals got inside an organization. Typically, they do so to launch a much larger attack such as ransomware. The Index also found that phishing was used in 41% of the attacks that X-Force remediated in 2021. That's a 33% increase from 2021. One of the biggest reasons threat actors are increasing phishing attacks is that all it takes is one employee to make a…

Top Security Concerns When Accepting Crypto Payment

From Microsoft to AT&T to Home Depot, more companies are accepting cryptocurrency as a way to pay for products and services. This makes perfect sense as crypto coins are a viable revenue source. Perhaps the time is ripe for businesses to learn how to receive, process and convert crypto payments into fiat currency. Still, many questions remain. How can you safely enable customers to pay with Bitcoin or other digital currency? What are the security risks that come with cryptocurrency? Let’s…

NFT Security Risks: Old Scams and New Tricks

The non-fungible token (NFT) boom has also led to some serious security incidents. For example, the number of suspicious-looking domain registrations with names of NFT stores increased nearly 300% in March 2021. To participate in an NFT marketplace, you must have an active cryptocurrency wallet. This exposes NFT holders to new risks as attackers can find ways into your crypto wallet through your marketplace account. As we’ll see, threat actors have even infiltrated NFT marketplace OpenSea’s Discord server posing as…