Hacktivists Use Automated Tools, Growing Reach to Target Government Organizations

Despite recent media attention surrounding nation-state hackers infiltrating government organizations and attempting to influence elections, the bulk of government-related cybercrime tends to be driven by less sophisticated and more ideologically-motivated campaigns carried out by hacktivist actors, according to a new report from SurfWatch Labs.

govriskchart
Government sector risk scores compared to the average for all sectors over the past year.

Government is the third most active sector when it comes to cybercrime, behind only information technology and consumer goods, and more than a third of the government CyberFacts collected by SurfWatch Labs this year have been related to hacktivist activity — far more than any other sector.

“The global reach of the Internet and social media along with the relative anonymity of cyber-attacks has provided hacktivists with a larger platform than ever to share their message, recruit new actors, and ultimately impact organizations,” noted the report, Cybercrime Gets Political: Automated Tools and Growing Reach Empowers Hacktivists.

It continued: “As a result, the most common cybercrime story in the government sector has involved websites and data being targeted by hacktivist groups resulting in service downtime, website defacement, and various types of information being stolen and publicly leaked.”

government-atep-4
SurfWatch Labs’ data shows that hacktivists have been the top trending actor category across many different government subgroups so far this year – in some cases appearing in more than two-thirds of CyberFacts.

Hacktivist-driven data breaches are not a new problem for the government sector. In 2013, the FBI warned that anonymous hacktivists using Adobe exploits were able to infiltrate agencies such as the U.S. Army, the Department of Energy, and the Department of Health and Human Services in order to steal sensitive information.

“It is a widespread problem that should be addressed,” the 2013 alert stated.

Three years later,  hacktivists remain as a top source of government-sector data breaches.

2016-09-27-govbreachactors
Hacktivists are the top trending known actor group associated with government data breaches so far in 2016.

Government agencies across the world have been targeted by hacktivists using well-known attack vectors such as SQL injections, social engineering and stolen credentials.

For example:

  • Shortly after Anonymous Philippines defaced the COMLEC website in protest of “questions and controversies” surrounding the country’s electoral process, LulzSec Pilipinas posted the entire COMLEC database online. The incident has been described as the largest government-related data breach in history – affecting more than 55 million people.
  • A hacker supporting Palestine published the names and personal information of FBI and Department of Homeland Security employees. The hacker said he first compromised the email account of a Department of Justice employee. Then he socially engineered access to the portal by pretending to be a new employee. Finally, he was able to find databases of employee information on the DOJ intranet.
  • The Anonymous #OpAfrica campaign led to several breaches including a one terabyte dump of information from Kenya’s Ministry of Foreign Affairs and International Trade. Kenya’s Ministry of Information and Communications Technology cabinet secretary Joseph Mucheru said the information was stolen due to a phishing attack that duped employees into clicking a link to change their credentials, which provided the hacktivist access to email accounts.
  • A hacker known as Hanom1960 breached several government agencies – including the Costa Rica Ministry of Culture and Foreign Affairs, the Columbia Ministry of Information Technologies and Communications, and Columbia’s Ministry of National Education – and subsequently leaked information on various government employees. “I see many mistakes in [their IT] systems,” the hacker told news outlets. “It is something that does not concern governments.”

government_hacktivistmicroeffectHacktivists are often characterized as graffiti artists or vandals that simply deface websites and cause other nuisance-level problems for organizations.

Those types of attacks are common, with SurfWatch Labs’ data showing that website downtime and website defaced are the most popular effects of hacktivism; however, the threats from hacktivists go beyond those simple attacks.

According to the report:

“Government officials noted in 2015 that the bulk of the cybercrime-as-a-service economy may be powered by as few as 200 individuals, yet those services put traditional cybercrime tools such as malware, botnets and DDoS attacks at the fingertips of a vastly larger pool of actors. … This trend, along with the large number of federal, state and local government agencies across the world, the global reach of hacktivist actors, and a never-ending series of political causes means that hacktivists have the ability, reach and will to cause harm to government organizations on a level never before seen.”

Hacktivists don’t have the resources of state-sponsored actors, but they are much more open about their attacks — often using public channels to coordinate attacks, gain media attention and recruit other actors to the campaign.

“This chatter can lead to valuable threat intelligence around what types of organizations are being targeted, how those attacks are impacting organizations and, ultimately, what can be done to better protect your organization,” the report concludes. “Monitoring hacktivist chatter and utilizing external cyber threat intelligence, along with your own internal data, can help to paint a full picture of the cyber risks facing your organization, determine what assets are at greatest risk, and inform where cyber defense efforts should be focused in the future.”

For more information, download the full report, Cybercrime Gets Political: Automated Tools and Growing Reach Empowers Hacktivists.

Learning from Cybercriminals: Using Public Tools for Threat Intelligence

Effective cyber threat intelligence is largely about gaining proper context around the risks facing your organization. As SurfWatch Labs chief security strategist Adam Meyer recently wrote, there are three pillars when it comes to evaluating those cyber threats: capability, intent and opportunity.

Threat_Triangle.pngThe first two, the capability and intent of threat actors, are mostly external aspects that you have no control over, but the third pillar, the opportunity for actors to exploit your organization, is something that can be controlled, evaluated and improved upon.

Malicious actors are relentless when it comes to finding information on that opportunity, and organizations need to use that same relentlessness when searching for potential weaknesses in their cybersecurity, according to a recent report from SurfWatch Labs.

“Knowing where attackers get their information and how they use it is an important piece of your overall cybersecurity strategy,” noted the paper, Top Sources of Information for Cyber Criminals: Where the Bad Guys Go to Conduct Research on Their Targets.

Over the past few months on this blog, we’ve profiled some of the top cyber threats and items for sale on various dark web marketplaces, but not all malicious activity occurs on this “underground web.” Much of it can be found wide out in the open — using simple tools and services that are available to anyone. Here are the top three public websites and tools used by malicious actors, as described in the paper, and how they can help those actors find the opportunity to attack your organization.

1. Shodan

Shodan was originally launched in 2009 by developer John Matherly and bills itself as “the world’s first search engine for Internet-connected devices.” This simple idea has grown from a basic list of IPs and ports to maps showing where devices are located to screenshots taken from these devices (including webcams, unsecured servers and workstations). The original focus for Matherly’s scans was to highlight the growing problem of the “Internet of Things,” but his research also uncovered industrial control systems, wide open computer systems, unsecured security cameras and more.

shodan

Researchers using Shodan frequently find publicly-exposed data that leads to breach notifications. Just one example is MacKeeper security researcher Chris Vickery discovering personal information from child tracking platform uKnowKids earlier this year.

“One of the uKnowKids databases was configured for public access, requiring no level of authentication or password and providing no protection at all for this data,” Vickery wrote. “There’s no way for me to know for sure how long this data was exposed to the public internet, although the information collected by Shodan.io suggests that the database had been up for at least 48 days.”

uKnowKids CEO Steve Woda reacted by describing Vickery as a hacker whose method “puts customer data and intellectual property at risk.” However, malicious actors can just as easily utilize Shodan to find opportunity for attacks.

As SurfWatch Labs’ paper summarized: “If it’s online, Shodan will find it. The lesson to be learned from this site, without a doubt, is secure something before it goes online.”

2. VirusTotal

VirusTotal describes itself as “a free service that analyzes suspicious files and URLs and facilitates the quick detection of viruses, worms, trojans, and all kinds of malware.” But that simple tagline masks a deeper set of capabilities.

vius-total

Security researchers have previously suspected that malicious actors use VirusTotal as a tool to help test and hone malware before sending it out in the wild, and in 2014 researcher Brandon Dixon confirmed those suspicions by discovering several hacking groups using the tool, including two nation-state groups.

Dixon said nation-state actors using a free online service to fine-tune their attacks was ironic and unexpected, but that speaks to the usefulness of VirusTotal.

“The power behind VirusTotal is how it adds and saves the metadata and behaviors of the files it analyzes,” noted SurfWatch Labs’ paper. “You can use the domain search to look at the IP history of the domain and get the current WHOIS for the domain, but VirusTotal will also show you a list of every time it detected something malicious on the site, as well as list all of the samples that attempted to communicate with the searched for domain.”

In addition to organizations using VirusTotal to help identify if they’ve been previously targeted, VirusTotal should be seen as a baseline site that can be used for detecting and analyzing suspicious and malicious files.

3. Your Own Company Website

The best way to get information about a particular company is often directly from the source: your own company website. Company websites can provide a treasure trove of information that can be leveraged by attackers to target a specific organization. This includes names of VIPs, email addresses of company executives and other employees, photographs, links to LinkedIn profiles and other social media, and more.

But beyond the surface level, there may be even more valuable information, as the paper explained:

Are you hosting any PDFs for people to download? Word documents, or PowerPoint presentations? Did you remember to remove potential metadata from those documents that could potentially contain additional names, email addresses, usernames, or software versions of the program used to create it? Some pretty simple Google searches (just type “site:yourpublicsite.com filetype:pdf” into the google search box) can reveal much more information that you may not have been aware you were “leaking.”

These types of leaks can lead to costly data breaches.

public

Free tools and services such as the ones described above provide malicious actors with valuable insight into the opportunity for cyber-attacks, and they are certainly one of the first places those actors turn to gather information on your organization. To make matters worse, all of this information can be discovered with minimal effort or expertise.

The good news is that those same tools can be used to gather cyber threat intelligence and to ensure that you are performing the same level of diligence as the threat actors who are trying to harm your organization.

Download SurfWatch Labs Top Sources of Information for Cyber Criminals: Where the Bad Guys Go to Conduct Research on Their Targets for more information.

Short Selling Vulnerabilities Latest in String of Stock Market Manipulation

Medical device company St. Jude filed a lawsuit yesterday against Muddy Waters and MedSec Holdings over a “false” report about cybersecurity issues in St. Jude’s cardiac devices. The August report caused the company’s stock to drop more than ten percent on the heels of those allegations and raised questions around a pending $25 billion deal to be acquired by Abbott Laboratories.

The heart of the issue is that MedSec Holdings, which discovered the alleged flaws, did not disclose them to St. Jude; rather, they took their findings to short-selling firm Muddy Waters in order to short St. Jude stock and turn a profit from the public disclosure.

Capture.PNG
MedSec contacted Muddy Waters with the proposal to short St. Jude stock after spending 18 months doing research and not generating any revenue, CEO Justine Bone said. Money made from shorting the stock will help finance development of secure medical device technology.

In its lawsuit, St. Jude said, “This insidious scheme to try to frighten and confuse patients and doctors by publicly disseminating false and unsubstantiated information in order to gain a financial windfall and thereby cause investors to panic and drive the St. Jude stock price down must by stopped and defendants must be held accountable so that such activity will not be incentivized and repeated in the future.”

The public battle has been at the center of an ongoing debate over the past two weeks — once again putting the issue of manipulating the stock market via cyber front and center.

Malicious Actors Profit From Stock Market

It’s no secret that malicious actors seek similar types of non-public information that can be used to leverage big profits in the stock market.

Perhaps the most famous recent case involves the theft of press releases from various newswire services. According to an August 2015 complaint filed by the Securities Exchange Commission (SEC), hackers gained access to the services, stole more than 100,000 press releases for publicly traded companies, and then used that information – often quarterly or annual earnings data – to reap over $100 million in unlawful profits.

As we noted in our 2015 Cyber Risk Report, the hackers worked with a network of traders to capitalize on the window between when a draft of a press release was provided and when it was made available to the public. In some instances that window was only a few minutes, but having that knowledge was extremely profitable, as the SEC complaint demonstrated.

2h2015_sec
By using non-public earnings information, the network of traders listed above were able to generate millions of dollars in profits through illegal trades.

Additionally, last summer reports of the hacking group Fin4 breaking into corporate email accounts to steal mergers and acquisitions data sparked the SEC to approach companies about possible breaches.

“The SEC is interested because failures in cybersecurity have prompted a dangerous, new method of unlawful insider trading,” John Reed Stark, a former head of Internet enforcement at the SEC, told Reuters.

Other cybercriminals have used less sophisticated methods to manipulate stock prices.

In July Gery Shalon, 32, and Ziv Orenstein, 41, were extradited from Israel and pled not guilty to charges that included a breach at JPMorgan Chase, which authorities described as the  “largest theft of customer data from a U.S. financial institution in history.” The stolen contact information was used to send deceptive communications in order to inflate stock prices, a practice known as pump and dump.

First, they would execute prearranged manipulative trades to cause the stock’s price to rise small amounts on successive days. Then they would send spam emails — sometimes millions a day — touting the stock. Finally, after artificially pumping up the price, they would dump their shares of the stock for huge profits.

A New White-Hat Shorting Strategy

While cyber-experts have long-pointed to the massive profits criminals can make from combining cyber-attacks with strategies such as shorting, the move towards white-hat hackers doing the same thing has created some concern.

MedSec CEO Justine Bone said she knows the approach they used will lead to criticism, but that it was the most powerful way to inflict pain on St. Jude over the company’s “negligent level of attention to cybersecurity.”

Although many companies have implemented bug bounties in an effort to encourage researchers and other hackers to disclose vulnerabilities in a responsible manner, those programs often don’t come with big payouts or spur the change desired by the person who disclosed the bug. Those players may attempt to copy the MedSec strategy — resulting in more profits and more public pressure to respond to alleged vulnerabilities. That gives yet another reason for investors to be concerned over potential cyber issues.

Medical device consultants Billy Rios and Jonathan Butts told Bloomberg that traders were clearly blindsided and scrambling over this new idea, having been inundated with requests from hedge funds, short sellers and other investors about the Muddy Waters report.

“This is almost like The Big Short,” Butts said. “Someone saw something that nobody else did.”

POS Breaches: Bankrupting Small Businesses and Impacting the Supply Chain

There’s a popular cybercrime statistic that has been vexing me for years, and if you read cybersecurity news regularly, I’m sure you’ve seen it cited a few dozen times as well:

60% of small businesses close their doors within six months of a cyber-attack.

I’ve always been skeptical of that bold statistic. As Mark Twain wrote in his autobiography, attributing the now famous quote to British Prime Minister Benjamin Disraeli, “There are three kinds of lies: lies, damned lies and statistics.” Sixty percent is incredibly high (and what percent of these companies would have failed anyway, cyber-attack or not?); nevertheless, I’ve always wanted to find the source of that data and delve into the stories behind that number.

I’ve largely failed on both of those fronts over the past few years.

First, the statistic is most often attributed in some vague way to either the National Cyber Security Alliance or the U.S. House Small Business Subcommittee on Health and Technology. In fact, National Cyber Security Alliance executive director Michael Kaiser did quote that statistic before the House Small Business Subcommittee on Health and Technology in December 2011, but he was actually citing a Business Insider article from three months prior. The Business Insider article is similarly vague, saying only that “about 60 percent of small businesses will close shop within six months of an attack” — but providing no other context to back up that assertion.

Second, my repeated attempts to find small businesses that have failed due to cyber-attacks — and are willing talk publicly about those failures — have come up mostly empty.

When Breaches Lead to Bankruptcy

All of this serves as a backdrop to the recent conviction of Roman Valerevich Seleznev, aka Track2, 32, of Vladivostok, Russia. Seleznev was convicted on August 25 of 38 counts related to hacking point-of-sale systems and stealing payment card information. According to trial testimony, Seleznev’s scheme led to more than $169 million in losses across 3,700 financial institutions.

Perhaps most interesting — at least when it comes to my ongoing quest to chronicle small businesses being put out of business by cybercrime — was this tidbit from the Department of Justice press release:

Many of the businesses [targeted by Seleznev] were small businesses, some of which were restaurants in Western Washington, including the Broadway Grill in Seattle, which was forced into bankruptcy following the cyber assault.

According to the indictment, Seleznev and others used automated techniques such as port scanning to identify vulnerable retail point-of-sale systems that were connected to the Internet and then infect those systems with malware.

“[Seleznev and others] hacked into, installed malware on, and stole credit card track data from, hundreds of retail businesses in the Western District of Washington and elsewhere,” the indictment stated. “[They] stole, in total, over two million credit card numbers, many of which they then sold through their dump shop websites … generating millions of dollars of illicit profits.”

Seattle’s iconic The Grill on Broadway was one of those small businesses to be hit by point-of-sale malware in 2010. The incident, along with other issues inherited from previous owners, led to the restaurant being closed in 2013.

“It became a target of a credit card number harvesting scheme that claimed a number of businesses on Broadway as victims,” the Seattle Gay Scene wrote at the time of the closing. “Several years of missed software updates played a significant role in the incident and [owner Matthew] Walsh and his team discovered this fact only a few months after purchasing the business. The effects were devastating to The Grill, generating massive amounts of negative publicity and drastically reduced revenue at the restaurant.”

The resources required to stay afloat were simply too much.

“In spite of what it may seem, we’re a very small business,” Walsh said. “We don’t have endless financial resources to keep us afloat like a chain restaurant or large corporation could.”

Recent Supply Chain Issues Affect POS Systems

The conviction of Seleznev over stolen payment card information and the re-emergence of The Grill on Broadway’s story comes during the same month that several point-of-sale vendors, including Oracle MICROS, have announced potential compromises — and a series of retailers and hotels have subsequently published data breach notifications.

Those breaches haven’t been explicitly connected, but several of the hotels to recently announce breaches have previously confirmed using MICROS products.

For example, Millennium Hotels & Resorts (MHR), which recently announced a data breach affecting food and beverage point-of-sale systems at 14 hotels, said it was notified by a third-party service provider about “malicious code in certain of its legacy point of sale systems, including those used by MHR.”

“The third party is a significant supplier of PoS systems to the hotel industry,” a spokesperson responded when SurfWatch Labs inquired about problems stemming from the supply chain. “It is aware of these issues. We are not disclosing the name.”

However, in 2008 MICROS Systems, now owned by Oracle, announced that Millennium Hotels & Resorts would be using MICROS “as the standard food and beverage point-of-sale solution for its 14 Millennium Hotel properties located in the United States” — so it’s possible there’s some connection between the breaches.

The same Russian group that hit MICROS has targeted at least five other cash-register providers, according to Forbes’ Thomas Fox-Brewster. Investigations are ongoing, but as we noted in our recent report, cybercrime is increasingly interconnected and compromises can quickly move down the supply chain, affecting everyone from small businesses to large enterprises.

If that 60% statistic is true, even partially, then it begs the question: will these recent breaches in the point-of-sale supply chain lead to more shuttered doors in the future?

And will we hear those businesses’ stories if it does happen? Or will they just become another vague statistic that we all continue to reference?

After Slow Start in 2016, Point-of-Sale Breaches Surging

Last week Eddie Bauer became the latest in a growing string of companies to announce a major point-of-sale-related breach. All 350 North American stores were affected by malware that may have siphoned off customers’ payment card information between January and July of this year.

Not all cardholder transactions were impacted, the company said, and the breach does not include any online transactions; however, the announcement comes during the same month that Oracle MICROS, HEI Hotels & Resorts and several other companies posted similar breach announcements.

The recent surge follows a comparatively quiet period over the first half of 2016, as this chart from our Mid-Year 2016 Cyber Risk Report highlights.

POS
Compared to the large number of POS breaches and chatter in 2014, the past year and a half has been relatively quiet — other than a spike in late 2015 tied to several different hotel breaches, the report said.

“This dip in discussion is accentuated by the extreme number of high-profile organizations affected by POS breaches in 2014, perhaps skewing the perception for what ‘normal’ levels of activity should be,” the report noted. “Point-of-sale breaches are not making as many headlines, but breaches so far this year have proven that for many organizations the associated costs are as high or higher than they have ever been.”

Revisiting that chart a month and a half later, it appears the activity level is now kicking up to match those high costs. SurfWatch Labs has collected more point-of-sale-related CyberFacts in August (through just 21 days) than any other month so far this year.

2016-08-22_POS_Chatter.png
The number of point-of sale CyberFacts collected by SurfWatch Labs has surged in recent months (data through August 21). HEI Hotels & Resorts is the highest trending POS-related target this month after announcing a data breach.

Oracle, Other Vendors Compromised

Adding to the concern around point-of-sale systems, Brain Krebs recently broke the news of a breach of hundreds of computer systems at Oracle, including a customer support portal for companies using Oracle’s MICROS point-of-sale credit card payment systems.

Sources said the MICROS customer support portal has been observed communicating with a server known to be used by the Carbanak Gang. That’s alarming since the gang is suspected be behind the theft of more than $1 billion from financial institutions in recent years.

“This breach could be little more than a nasty malware outbreak at Oracle,” Krebs wrote. “However, the Carbanak Gang’s apparent involvement makes it unlikely the attackers somehow failed to grasp the enormity of access and power that control over the MICROS support portal would grant them.”

The investigation is ongoing, and Oracle so far has not provided customers or media outlets with many answers.

To make matters worse, Forbes’ Thomas Fox-Brewster reported that several other cash register suppliers besides MICROS have been breached recently.

“It now appears the same allegedly Russian cybercrime gang has hit five others in the last month: Cin7, ECRS, Navy Zebra, PAR Technology and Uniwell,” he wrote. “Together, they supply as many as, if not more than, 1 million point-of-sale systems globally.”

Hotels Remain Top Trending POS Target

In our mid-year report, the “Hotels, Motels and Cruiselines” subgroup of Consumer Goods dominated the chatter around point-of-sale breaches, and not much has changed in the two months since that report. In fact, nearly 42% of all the point-of-sale CyberFacts collected by SurfWatch Labs so far this year have fallen into that group.

2016-08-22_POS_Groups
More than 60% of SurfWatch Labs’ point-of-sale related CyberFacts collected this year fall into either the Hotels, Motels and Cruiselines or Restaurants and Bars groups.

The top trending point-of-sale target this month is HEI Hotels & Resorts, which announced a breach involving 20 hotels on August 12. The malware was discovered in June on point-of-sale systems used at restaurants, bars, spas, lobby shops and other facilities, according to Reuters. Twelve Starwood hotels, six Marriott International properties, one Hyatt hotel and one InterContinental hotel were impacted.

If those names sound familiar, it’s because several of them have already made news for data breaches of late, including Hyatt in December 2015 and Starwood in January 2016.

Other data breaches this year involving hotels include Kimpton Hotels, Hard Rock Hotel & Casino Las Vegas, Rosen Hotels & Resorts and the Trump Hotel Collection.

2016-08-22_POS_Groups_ITT

Although the various incidents that have been announced in recent weeks have not been explicitly connected by either researchers or law enforcement, the breach notice from Eddie Bauer did signify that other organizations have been targeted with a similar campaign.

“Unfortunately, malware intrusions like this are all too common in the world that we live in today,” the company wrote. “In fact, we learned that the malware found on our systems was part of a sophisticated attack directed at multiple restaurants, hotels, and retailers, including Eddie Bauer.”

Other experts such as Gartner fraud analyst Avivah Litan have speculated that the breach at Oracle “could explain a lot about the source of some of these retail and merchant point-of-sale hacks that nobody has been able to definitively tie to any one point-of-sale services provider.”

At the moment many questions remain, but if these investigations lead to the discovery of further compromises, expect to see more breach announcements and more payment card information being sold on Dark Web markets in the months to come.

Typosquatting: Easy Attack Vector That Produces Results

Every week here at SurfWatch Labs our team of threat analysts write about new vulnerabilities, malware developments and cyber-attacks.  One attack vector that is not mentioned very frequently but can be a significant threat for organizations and consumers alike is a technique called typosquatting.

Typosquatting is an attempt to trick users into thinking they have landed on their desired website, but in reality the user has landed on a website with a similar looking domain name that is controlled by cybercriminals. It’s an old technique, and security-conscious organizations often try to secure those domain variations that arise from typos.

However, a study last year described how companies remain vulnerable to typosquatting and found that most organizations do very little to protect their customers from the threat.

Key findings from the study:

  • Few trademark owners protect themselves against typosquatting by defensively registering typosquatting domains for their own domains.
  • The study found that 95% of the most popular 500 websites researched were targeted with typosquatting.
  • Hackers are increasingly targeting longer domains.
  • Some companies secure potential typosquatting domains but then choose not to renew them, leaving them vulnerable.

TypoSquatting Attack Example

A great example of a typosquatting attack was used against the popular online first-person shooter game Counter-Strike: Global Offensive. The hackers set up a convincing spoof, tricking gamers into believing they were on a legitimate site for the game. The fake site was listed as csgoloungcs.com, while the legitimate site is csgolounge.com.

Not only were visitors of the fake site tricked into sharing their login credentials, a Trojan downloader was pushed on them, leading to malware infections.

Another example found malicious actors taking advantage of the .om top level domain. Earlier this year, Netflix users who mistyped the address as netflix.om were redirected to a fake Flash update page.

Typosquatting is one example of the many opportunistic type of threats facing organizations. It doesn’t require sophisticated techniques, and it’s an easy way to leverage popular brands in order to entrap customers who aren’t aware of such scams.

Typosquatting scams can lead to a variety of consequences for users — from account takeover to identity theft — and those consequences can easily spill over to the organizations being impersonated in the form of disgruntled customers, bad press, or having to deny a breach when stolen credentials are put up for sale on the Dark Web.

All that trouble can be largely avoided by being vigilant about identifying common typographical mistakes related your organization’s domains and purchasing them to keep them out of malicious actors’ hands.

IcyEagle: A Look at the Arrest of an Alleged Dark Web Vendor

Last month Aaron James Glende, 35, was arraigned in U.S. District Court in Atlanta on charges related to selling stolen bank account information on the Dark Web market AlphaBay. According to the indictment, Glende operated under the alias “IcyEagle” and began advertising his criminal services in late 2015.

Although the exact picture of how law enforcement managed track down and identify Glende remains unclear, the details released so far provide an interesting behind the scenes view of the cybercrime-related postings we often highlight on this blog.

IcyEagle_SunTrust
IcyEagle listed these high-balance SunTrust Bank accounts for sale on AlphaBay in May 2016. He sold similar items to an undercover FBI agent in March and April 2016.

SurfWatch Labs has observed IcyEagle selling information related to a variety of companies over the past 10 months, but the June 28 indictment mentions only one company by name, SunTrust Bank.

On multiple dates in March and April 2016, a Federal Bureau of Investigations (“FBI”) agent in the Northern District of Georgia, acting in an undercover capacity, accessed the AlphaBay website. While on the website, the agent purchased SunTrust account information from Icy Eagle using Bitcoin. A review of the information purchased from IcyEagle confirmed that it contained usernames, passwords, physical addresses, email addresses, telephone numbers, and bank account numbers that belonged to five different SunTrust Bank customers.

IcyEagle has listed SunTrust Bank accounts with a variety of balances this year, ranging from $250,000-$500,000 (selling for $229.99), to $100-$500 (selling for $9.99).

He also sold a 6-page guide on how to best cash out SunTrust Bank accounts, which includes sections on routing numbers, background checks, Bitcoin, and other tips.

IcyEagle_SunTrustGuide
IcyEagle sold guides on how to cash out compromised accounts, including SunTrust Bank accounts. As with many listings on Dark Web markets, guides on using those items or services are readily available.

“I bring you freshly hacked Sun Trust Bank Account Logins,” read one posting for SunTrust Bank accounts with balances between $30,000 and $150,000. “The accounts are notorious for having weak security.”

According to postings viewed by the FBI, IcyEagle sold at least 11 of these high-balance SunTrust Bank accounts and 32 of the lower-balance accounts.

Dozens of other listings not-related to SunTrust Bank were also posted by IcyEagle and likely sold this year, although those were not listed in the recent indictment. 

IcyEagle_Amazon
Amazon is one of the most popular companies tied to IcyEagle in SurfWatch Labs’ data, based on the number of listings we have observed on AlphaBay.

IcyEagle sold hacked Amazon gift balances for around one-tenth of the total balance. Other accounts for sale generally ranged from $2.99 to $14.99, depending on the type of account. These included email logins, dating website logins, customer reward program logins, logins for various financial services and more.

How was IcyEagle Caught?

An undercover officer purchased stolen bank account information from IcyEagle in March and April 2016, according to the indictment. Interestingly, Glende was also arrested by local police for selling drugs around the same time. A tip from U.S. Postal Inspectors led to police officers finding a “trove” of drugs at his Minnesota home in March.

“According to police, Postal inspectors reported finding packages connected to Glende that contained prescription pills,” wrote the Winona Post. “Officers executed a search warrant of Glende’s home on Friday, March 11, and reportedly found two U.S. Postal Service packages ready to be sent that contained the prescription narcotics Valium, Xanax, and oxycodone. Officers reportedly found a trove of other drugs at Glende’s home: nearly 600 Xanax pills, more than a dozen dextroampethamine capsules, 138 oxycodone pills, nearly 50 Valium pills, marijuana, and marijuana wax.”

The indictment states that IcyEagle began advertising his criminal services by early November 2015. SurfWatch Labs’ data matches these allegations, with our threat intelligence analysts first observing several listings by IcyEagle in October 2015. New listings continued to be posted until the end of May, shortly before his arrest. 

2016-08-10_IcyEagleActivity.png
SurfWatch Labs has been observing IcyEagle listing cybercrime-related items on AlphaBay Market since October 2015.

It’s unclear how — or even if — those two events are linked, but shortly after that drug-related arrest the FBI appears to have begun targeting IcyEagle’s postings on AlphaBay. We can speculate that after U.S. Postal Inspectors tied Glende to selling prescription drugs, the search warrant and subsequent investigation may have revealed evidence leading law enforcement to AlphaBay and IcyEagle — or vice versa. Either way, Glende is charged with performing cybercrime-related activities including five counts of bank fraud, four counts of aggravated identity theft, and one count of access device fraud.

Law enforcement officials continue to tout the arrests of alleged cybercriminals such as Glende as a sign that they will hold bad actors accountable for their actions despite the difficulties associated with such a task.

“The threat posed by cyber criminals is a persistently increasing problem for everyday citizens here in the U.S. and abroad,” said J. Britt Johnson, Special Agent in Charge, FBI Atlanta Field Office, in a press release. “This investigation and resulting arrest clearly illustrates that the FBI, however, will not cease in its effort to identify, locate, arrest and seek prosecution of these criminals regardless of how deep in the digital underground they reside.” 

IcyEagle was just a drop in the bucket when compared to the thousands of pieces of Dark Web threat intelligence SurfWatch Labs analysts have recently observed. Nevertheless, cases like this serve as an important reminder of the insight that can be gained by watching these markets — not just for law enforcement, but for the companies that bear the brunt of this malicious cybercrime activity.

Payment Transactions Face New Data Breaches and Exploits

The last few weeks have not been kind to businesses and customers concerning payment transactions and digital currency. Several point-of-sale systems and digital wallet services have come under fire for data breaches and potential financial theft — not to mention the recent theft of $68 million worth of bitcoin.

The most wide-reaching event may be the breach at software company Oracle Corp, which was reported by Brian Krebs on Monday. A Russian cybercrime group appears to be behind an attack that saw the compromise of hundreds of computers system, including a customer support portal for Oracle’s MICROS point-of-sale credit card payment systems.

This could be a potentially huge breach, as more than 330,000 cash registers around the world utilize Oracle’s MICROS point-of-sale system. In 2014, the company said that about 200,000 food and beverage outlets, 100,000 retail sites, and 30,000 hotels used the software.

It is currently unknown how many organizations were affected by the breach or how long the breach took place. The investigation is ongoing, but potential ties to the Carbanak Gang have raised the level of concern. Oracle did tell Brian Krebs that the company “detected and addressed malicious code in certain legacy MICROS systems,” and that Oracle asked customers to reset their MICROS passwords.

Digital Wallets Face Scrutiny

At last week’s Black Hat conference, a security researcher presented on a flaw in the mobile payment system Samsung Pay. Samsung Pay allows customers to save payment cards on a digital wallet, providing users the option to select the payment card of their choice with the added security of a PIN or fingerprint scan to complete a purchase.

Security expert Salvador Mendoza discovered several problems with Samsung pay, including static passwords used to protect databases, weak obfuscation, and comments in the code. Mendoza also discovered issues with the tokens that are used to complete transactions. Cybercriminals could potentially predict future tokens from studying previous tokens used to make fraudulent transactions.

“Samsung Pay has to work harder on the token’s expiration date to suspend it as quickly as possible after the app generates a new one, or the app may dispose of the tokens which were not implemented to make a purchase,” Mendoza explained. “Also, Samsung Pay needs to avoid using static passwords to ‘encrypt’ its files and databases with the same function because eventually someone will be able to reverse it.”

Samsung responded to Mendoza’s claims by saying “reports implying that Samsung Pay is flawed are simply not true.”

However, in a separate document Samsung did admit that “skimming” a token is possible, although extremely difficult.

“Samsung Pay’s multiple layers of security make it extremely difficult to make a purchase by skimming a token,” the company wrote. “This skimming attack model has been a known issue reviewed by the card networks and Samsung pay and our partners deemed this potential risk acceptable given the extremely low likelihood of a successful token relay attack.”

Samsung Pay isn’t the only digital wallet in the news for potential cybersecurity issues.  Venmo — a digital wallet service that allows users to interact with friends by sending money, making purchases, and sharing payments — made headlines recently for flaws that could potentially lead to malicious purchases.

A flaw in an optional SMS-based feature could allow a criminal to easily steal money from people’s accounts, according to researchers. Because Venmo allows users to charge friends through shared bill pay, that friend has to authorize the charge before payment is made. A hacker with physical access to a Venmo user’s phone could steal money from another user’s account by replying to a notification text message with a provided 6-digit code. A feature in Siri that allows users to reply to text messages from locked devices along with the iOS text message preview feature make this attack possible.

“A hacker could have sent a payment request to a targeted user, and if they had access to the victim’s locked device, they could have used Siri to send the approval code displayed on the screen, ” said Eduard Kovacs of SecurityWeek. “The maximum amount of money an attacker could have stolen from one user was $2,999.99 per week, which is the weekly limit set by the developer.”

Keeping Payments Safe

As we’ve highlighted on this blog and in recent threat intelligence reports, high-profile payment-related breaches aren’t at the forefront of cybercrime in the way they were several years ago. However, recent events prove that these payment systems — traditional point-of-sale systems, digital wallets and digital currencies — can lead to significant direct losses as well as brand damage and other consequences from the negative press generated by discovered vulnerabilities.

As SurfWatch Labs’ Chief Security Strategist Adam Meyer recently wrote, cybersecurity is largely about identifying and removing opportunity for malicious actors to do bad things — either directly or indirectly.  There are clear best practices that can be utilized by both businesses and customers to help protect sensitive payment data. Unfortunately, data is only as safe as the methods used to protect it.

Cybercriminals are constantly coming up with new methods and tricks to crack software and trick people into divulging their sensitive information. Cyber threat intelligence can help organizations remain mindful of the many new and evolving threats, identify their weaknesses, and deploy safeguards to protect data — whether that is payment-related data or other sensitive information.

 

Top Dark Web Markets: TheRealDeal, Paranoia and Zero-Day Exploits

In trying to demystify the Dark Web, we’ve talked about the customer-friendly features of markets, the hand-holding nature of cybercrime-as-a-service, and the secure payment options that can protect anonymous buyers.

As we turn our attention to the exploit-centric TheRealDeal Market, it gives us a chance to examine an aspect of the Dark Web that isn’t so rose-colored: the paranoia that runs deep for many buyers, sellers and market operators.

A Quick Look at TheRealDeal Market

Of the many Dark Web markets, TheRealDeal Market has perhaps the most interesting backstory. While other markets focus on things such as drugs and stolen payment card information, TheRealDeal Market launched in 2015 with a focus on code — from zero-day exploits to known vulnerabilities to source code. This led to stories in high-profile outlets such as Wired, which described TheRealDeal as “a new marketplace [that] hopes to formalize that digital arms trade.”

Shortly after making those headlines, several members were arrested, the site went offline, came back online for a short period, and then disappeared again. Finally, half a year later, it relaunched in December 2015.

TheRealDeal9
Exploits such as this alleged zero-day for ecommerce software are frequently listed on TheRealDeal Market.

The main reason for the long downtime was “paranoia,” as the site admin put it in an interview. That paranoia was grounded in real world events.

On July 13, 2015, the popular cybercrime forum Hell was shut down after its administrator, Ping, was arrested. A few days later — on July 15, 2015 — the FBI announced the dismantling of a dark web forum known as Darkode, which U.S. Attorney David J. Hickton described as “a cyber hornets’ nest of criminal hackers ” and “one of the gravest threats to the integrity of data on computers in the United States and around the world.” The coordinated law enforcement effort, known as Operation Shrouded Horizon, led to 70 Darkode members and associates across 20 countries being charged, arrested or searched.

As DeepDotWeb reported, those arrests tied up several members of TheRealDeal Market team.

“What I can say is that most of the original team is not with us at the moment,” TheRealDeal admin said in December. “Currently, at least for the time being, the market will be under the management of me (identified in support as admin S.P.), an old vendor that has stuck with us from the beginning, and a couple of trustworthy people from other darknet communities. I can also add that the main reason of the last down time was paranoia, if it turned out to be justifiable or not, I cannot say.”

That paranoia tends to run throughout all dark web markets — paranoia of law enforcement, paranoia of exit scams, paranoia of other users. As one drug vendor from the now-defunct Evolution Market said in a 2014 interview, “In this business it’s always better to be too paranoid than not paranoid enough.”

Feeding into that paranoia is the fact that the main administrator appears to have vanished recently and support has stopped replying to messages, at least according to one popular vendor.

“This [is] very strange by just leaving the market like this without any management or any notice for leaving,” the vendor told Motherboard in an online chat.

Yesterday, TheRealDeal Reddit account said the reason for the absence was an accident.

“Admin not dead, just almost,” the account wrote. “The only guy with the actual keys to the kingdom had a small accident. … More coming soon.”

What’s For Sale on TheRealDeal Market?

Since its December 2015 relaunch, TheRealDeal Market has once again been making national news. Most recently was a vendor selling a 200 million-strong database of alleged Yahoo user credentials, most likely stolen in 2012, for 3 bitcoin (around $1800). A Yahoo spokesperson said the company is aware of the listing and is investigating whether the data is legitimate.

The same vendor has recently sold massive databases of credentials from LinkedIn and MySpace.

TheRealDeal2
A posting from TheRealDeal Market claiming to contain 200 million user passwords for Yahoo.

TheRealDeal Market sparked another national story this summer when a different vendor began offering a series of healthcare databases for sale. That actor was able to use the media — along with initial price tags in the hundreds of thousands of dollars — to generate a significant amount of attention around the postings, his or her alias, and TheRealDeal Market. A half dozen databases have since been posted ranging from 23,000 records to 9.3 million records.

One of the more recent postings is from a healthcare organization in Fairview, Illinois.

TheRealDeal7.jpg
The seller claimed he or she was able to access various healthcare databases due to a zero-day vulnerability “within the RDP protocol that gave direct access to this sensitive information.”

“[The data] was retrieved from an accessible internal network using account credentials that were garnered through the token impersonation of an employee,” the listing reads. “First stage access was accomplished using RDP 0day.”

Although various stolen databases have generated most of the media attention around TheRealDeal Market, code-related items are a staple of the market — and one of the reasons it was founded.

“We actually tried selling such information and codes ourselves at some point [on established marketplaces] but it seems that all people want on those markets is credit cards and tutorials on how to make money with credit cards,” an admin said in an interview in April 2015. “The problem is that 90% of these dealers are scammers. People with a lot of experience can always do their best to determine if what they are buying is real based on technical information and demos but some of these ‘vendors’ are very clever and very sneaky. We decided it would be much better if there was a place where people can trade such pieces of information and code combined with a system that will prevent fraud and also provide high anonymity.”

The past month SurfWatch Labs has observed various alleged zero-day exploits for sale on TheRealDeal Market.

These include a listing claiming “a remote code execution that allows installation of any APK file on any Android phone that has [a certain gambling application] running.”

TheRealDeal4
The alleged zero-day exploit is selling for 12 bitcoin (around $7000).

There is also a posting claiming a local privilege escalation zero-day that will “go from user to SYSTEM in a few lines of code.”

TheRealDeal6
This alleged local privilege escalation zero-day is also selling for 12 bitcoin (around $7000).

Then there is an alleged zero-day in a popular messaging app, which can lead to denial of service.

TheRealDeal5
This denial-of-service exploit for a popular app is listed for 7.5 bitcoin (around $4,500).

In addition to zero-days and known exploits, there is also a variety of source code and other listings that can be found on the TheRealDeal Market.

For example, a recent listing claims to be selling information stolen from a large HL7 developer located in the United States.

“In addition to the source code for the HL7 Interface Engine software, the private keys for signing the code will also be included as well as the licensing database that entails a full record of all clients and their status information,” the listing reads. “There are many legitimate and nefarious uses for this exclusive package offer. You are only limited by your imagination.”

TheRealDeal1
This source code, signing keys and licensing database from a U.S. HL7 software developer comes with a hefty price tag of 40 bitcoin (around $23,000).

Another listing offers an enterprise code signing certificate.

“No timewasters please, if you don’t know why this is so expensive or what to do with it  — don’t bother,” the seller wrote.

TheRealDeal3.jpg
This enterprise code signing certificate is listed for 15 bitcoin (around $9,000).

These are just a sampling of the many recent listings on TheRealDeal Market.

Although TheRealDeal Market may not be as popular as AlphaBay or other markets that we’ve profiled over recent months — which tend to be dominated by things such as illegal drugs, hacking tutorials, payment card information and stolen credentials — TheRealDeal Market has managed to frequently make headlines for the types of information sold there, connections to other high-profile arrests, and now, the recent disappearance of the current admin.

OurMine Hacking Group Trending, What Are They After?

As we mentioned in a previous post, hacktivism activity has been down in 2016 — with the exception of Anonymous. However, there is a new hacktivist group that has been showing up in SurfWatch Labs’ data — OurMine.

Over the last two months, OurMine has been the top trending hacktivist group.

2016-08-02_hacktivist

OurMine made multiple headlines over the past month after successfully hacking the LinkedIn and Pinterest accounts of billionaire Facebook CEO Mark Zuckerberg. The hack provided some embarrassment for Zuckerberg, as it was discovered that the password he used for both accounts was “dadada.”

The group’s latest target was the CEO of Pokemon GO, John Hanke. OurMine hacked into Hanke’s Twitter account, saying that the hack was “for Brazil.”

Here are the top trending targets associated with the OurMine hacking team over the last two months.

2016-08-02_ourminetargets

What is OurMine After?

What separates OurMine from other hacktivists is their claim for hacking. In each of the group’s attacks, they claim they are a security firm that is testing their target’s security, and have even gone as far to say they were going to offer security services to their victims. The hacking group even has a website advertising their services.

OurMine has shown an aptitude for hacking. In several of their hacks — like Mark Zuckerberg’s social media — they were able to take advantage of a weak password to compromise the account. In other attacks — such as the attack against Google’s CEO Sundar Pichai’s Quora account — they have been able to exploit website platform vulnerabilities.

The group isn’t only after high profile businessman. OurMine has also targeted Minecraft player accounts, defaced websites like TechCrunch, and completely disabled the servers of HSBC bank.

It appears that all of these attacks are used as a method to promote their services. OurMine has yet to cause significant damage with any of their attacks other than a minor nuisance. Is this group’s supposed white hat hacking attempts really an effort to promote their security services and point out security weaknesses for companies? Only time will tell.