Deep learning: An explanation and a peek into the future

Deep learning is one of the most advanced forms of machine learning, and is showing new developments in many industries. In this article, we’ll explain the concept and give some examples of the latest and greatest ways it’s being used.

What is deep learning?

There have been many attempts at creating a definition of deep learning.

As we’ve explained in the past, machine learning can be considered as a sort of offspring of artificial intelligence. In the same way, you can view deep learning as a further evaluated type of machine learning.

According to Wikipedia:  Deep learning (also known as deep structured learning) is part of a broader family of machine learning methods based on artificial neural networks with representation learning. Learning can be supervised, semi-supervised or unsupervised.

While that definition does give us some clues on what we are looking at, it deserves an explanation of some of the terms used.

Artificial neural networks (ANNs) are computerized networks that mimic the behavior of biological communication nodes. What makes biological neural networks different from other artificial networks is that they are dynamic and analog. That not only makes them more flexible, but it also makes them harder to mimic in an artificial neural network.

Representation learning or feature learning is a set of techniques that allows a system to automatically discover the representations needed for feature detection or classification from raw data. In other words, representation learning is a way to extract features from unlabeled data by training a neural network

How is deep learning more advanced?

Basic machine learning methods are becoming better at what they were designed for at an impressive speed. But they still need human guidance from time to time. For example, when users notice that the algorithm has accepted a false statement as true. In such a case, the predictions made by the algorithm become worthless and the situation needs to be corrected.

Deep learning uses multiple layers which allows an algorithm to determine on its own if a prediction is accurate or not. As we all know, you can sometimes reach an accurate conclusion based on false facts. A deep learning model will typically be designed to analyze data with a logic structure and do that in a way that’s very similar to how a human would draw conclusions. This layered approach results in a method that is far more capable of self-regulated learning, much like the human brain.

The obvious warning here is that not every human brain is capable of following the rules of logic and while we perfect the mimicry, we may introduce the same weaknesses that exist in biological brains. Of course, deep learning machines are capable of processing a lot more input than humans can at this point, which is why big data and deep learning often go hand in hand.

Examples of deep learning

Machine learning and, more specifically, deep learning already have proven their worth in some use cases and we can expect more improvements in these fields.


Traffic analysis: Predictions about which roads and motorways are acting as a bottleneck and how the flow can be optimized with a minimum of investments. For example, whether it will prove to be useful to add an extra lane to that highway or whether it will just create the same problem a few miles further ahead.

Transportation automation: In transport, the shortest route is not always the fastest. A delivery route can be optimized by time of arrival at certain delivery addresses, which is something that can be done by deep learning.

Finding cures: Deep learning neural networks can help in structuring and speeding up drug design. Researchers have enhanced deep learning for drug discovery by combining data from a variety of sources.

Market analysis: Combining machine learning with your data can provide insight into which leads prove to give you the highest success rate. However, given that you need a relatively big dataset, this may not be interesting for smaller organizations lest it may lead to self-fulfilling prophecies.


Speech recognition: Apps that listen to voice commands can learn to understand their user better over time. This can help to overcome the returning annoyance about voice assistants that misunderstand or not understand the user at all.

Gesture recognition: One of the latest additions in the area of machine learning deals with recognizing gestures. The signals that are emitted from sensors are able to detect emotions by energy, time delay, and frequency shift.

Deepfakes: For good or bad, further analysis of facial expressions and voice patterns can provide the data for the next step in creating more convincing deepfakes. By better understanding human behavior, it will become easier to mimic and provide more convincing results.


Smartphone cameras: These small cameras have to make up for the limitations set by their size in order to come close to the picture quality made by dedicated cameras. Machine learning algorithms do several things to improve and enhance the smartphone’s picture quality.

Targeted advertising: To minimize the number of advertisements the public have to watch, and to optimize the effectiveness of those advertisements, deep learning can be used to provide targeted advertising and make sure the aim is at the most suitable demographic for your product.

These are just some examples. You can probably come up with more if you look around you and see how software has taken over a lot of tasks that required human brains in the past.

The use of machine learning has also made things possible that were impossible before. For example, Google built a system to guard the rainforest. The company built a solution based on an open source platform for machine learning that uses audio to detect sounds of chainsaws and logging trucks to understand if any if an illegal activity is occurring. The machine learning solution takes into account various artificial intelligence techniques to ensure it is correctly detecting any destruction taking place.

The cybersecurity industry

We’ve already talked at length in another blog about how artificial intelligence and machine learning may impact cybersecurity. Some of these changes are already taking form and others are well on their way to being developed, but as we move forward there are bound to be changes. Especially in an industry that is involved in an arms race that entices both sides to stay one step ahead of the other.

The post Deep learning: An explanation and a peek into the future appeared first on Malwarebytes Labs.

Baltimore gets hit by ransomware again, the schools this time

All Baltimore County Public Schools closed Wednesday after the school system was hit with a ransomware attack, according to officials.

Baltimore County Public Schools superintended Dr. Darryl Williams stated:

“This morning, we decided to close all BCPS schools and offices in order to access and limit the impact of the attack.”

For those unfamiliar with the Baltimore City Schools organization, the attack affected some 175 schools, programs, and centers, over 115,000 students, and over 18,000 employees.

In May of last year a RobbinHood ransomware attack paralyzed Baltimore’s City government, shutting down online systems for paying water bills and other services.

Measures taken by Baltimore City Schools

Since the attack also took down the official website, management is providing updates over social media channels.

Via their Twitter account, Baltimore County Public Schools announced the schools and offices to be closed on Wednesday, November 25. Later they added Monday, November 30, and Tuesday, December 1, to focus on identifying and addressing student and staff device needs so that instruction can continue.

On their Facebook account they urged people not to log into BCPS devices or systems at this time. They also reassured the public that they are doing their best to address the ransomware attack. Local, state, and federal law enforcement agencies are investigating.

Also via Twitter they asked students learning virtually on Wednesday to only use City Schools-issued laptops or devices. Those without those issued devices were granted an excused absence. BCPS-issued Chromebooks were not impacted by the cyberattack.

The Teachers Association of Baltimore County is telling parents to leave computers off and not turn it on until they hear back from BCPS.

Superintendent Darryl Williams said there is no timeline for when school will resume. According to school officials, the network issue has affected the district’s website, email system and grading system. Until the problem is resolved, students will be unable to attend school.


The county police have been in contact with the FBI Baltimore field office. Baltimore County Police Chief Melissa Hyatt declined to provide any specifics of the criminal probe, since they still are in the preliminary steps of that investigation.

Hyatt did not reveal whether the authorities have communicated with the hackers and the school system said it has had no direct or indirect contact with the hackers.

While it is important to investigate ransomware attacks, most of these investigations may not lead to the apprehension of the attackers. They could, however, reveal how the attackers got in and whether they left any backdoors for future use behind.

Ransomware and education

The educational system and many of its elements are targets for cybercriminals on a regular basis. While education is a fundamental human right recognized by the United Nations, the financial means of many schools and other entities in the global educational system are often limited.

You’d think there are more profitable targets for cybercriminals than education. Technology and finance, for example, have exponentially bigger budgets that could be tapped into via large ransom demands. But cybercriminals are opportunistic: If they see an easy target ripe with valuable data, they’re going to take advantage. Why spend the money and time developing custom code for sophisticated attack vectors when they can practically walk through an open door onto school networks?

With some ransomware gangs now creating extra leverage by threatening to publish exfiltrated data, criminals may well see schools as an easy target—expecting them to pay the ransom through fear of finding students’ and teachers’ personally identifiable information (PII) published online.

The timing for an attack to take out the network information systems, could not have been worse while the school system continues to operate online only, with all in-person classes delayed, as a result of the coronavirus pandemic. Possibly these circumstances could have provided the way in for the attackers. Hopefully the investigation will reveal how it happened.

Stay safe, everyone!

The post Baltimore gets hit by ransomware again, the schools this time appeared first on Malwarebytes Labs.

German users targeted with Gootkit banker or REvil ransomware

This blog post was authored by Hasherezade and Jérôme Segura

On November 23, we received an alert from a partner about a resurgence of Gootkit infections in Germany. Gootkit is a very capable banking Trojan that has been around since 2014 and possesses a number of functionalities such as keystroke or video recording designed to steal financially-related information.

In this latest campaign, threat actors are relying on compromised websites to socially engineer users by using a decoy forum template instructing them to download a malicious file.

While analyzing the complex malware loader we made a surprising discovery. Victims receive Gootkit itself or, in some cases, the REvil (Sodinokibi) ransomware. The decision to serve one or the other payload happens after a check with the criminal infrastructure.

Gootkit attacks observed in Germany

Security researcher TheAnalyst was the first to publicly identify an active campaign in November using a sophisticated loader that was eventually attributed to Gootkit, a banking Trojan not observed in the wild for some time. Germany’s Computer Emergency Response Team DFN-CERT later confirmed that German users were being targeted via compromised websites.

Around the same time, we started receiving reports from some of our partners and their ISPs about Gootkit-related traffic. We were able to confirm Gootkit detections within our telemetry that were all located in Germany.

Figure 1: Gootkit infections in Germany in the wake of the campaign

After a couple of days, we remediated over 600 unique machines that had been compromised.

Fake forum template on hacked websites

The initial loader is spread via hacked websites using an interesting search engine optimization (SEO) technique to customize a fake template that tries to trick users to download a file.

The template mimics a forum thread where a user asks in German for help about a specific topic and receives an answer which appears to be exactly what they were looking for. It’s worth noting that the hacked sites hosting this template are not German (only the template is); they simply happen to be vulnerable and are used as part of the threat actor’s infrastructure.

Figure 2: Compromised site loads decoy template to trick victims

This fake forum posting is conditionally and dynamically created if the correct victim browses the compromised website. A script removes the legitimate webpage content from the DOM and adds its own content (the template showing a link to a file to download).

Figure 3: A view of the HTML code behind the decoy template

There is a server-side check prior to each visit to the page to determine if the user has already been served the fake template or not, in which case the webserver will return legitimate content instead.

Fileless execution and module installation

The infection process starts once the victim executes a malicious script inside the zip archive they just downloaded.

Figure 4: Malicious script, heavily obfuscated

This script is the first of several stages that leads to the execution of the final payload. The following diagram shows a high level overview:

Figure 5: Infection flow

Stage 1 – The first JavaScript

The first JavaScript is the module that has to be manually executed by the victim, and it has been obfuscated in order to hide its real intentions. The obfuscation consists of three layers where one decodes content for the next.

The first stage (a version with cleaned formatting available here) decodes the next element:

Figure 6: First stage script

The decoded output is a comma-separated array of JavaScript blocks:

Figure 7: Randomly-named registry key to store payload

There are four elements in the array that are referenced by their indexes. For example, the element with the index 0 means “constructor”, 1 is another block of JavaScript code, 2 is empty, 3 is a wrapper that causes a call to a supplied code.

Block 1 is responsible for reading/writing registry keys under “HKEY_CURRENT_USER\SOFTWARE\<script-specific name>”. It also deobfuscates and runs another block of code:

Figure 8: Third JavaScript layer

This fragment of code is responsible for connecting to the C2. It fetches the domains from the list, and tries them one by one. If it gets a response, it runs it further.

The above script is the first stage of the loading process. Functionality-wise it is almost identical in all the dropped files. The differentiation between the variants starts in the next part, which is another JavaScript fetched from the C2 server.

Stage 2 – The second JavaScript (downloaded from the C2)

The expected response from the server is a decimal string, containing a pseudorandom marker used for validation. It needs to be removed before further processing. The marker consists of “@[request argument]@”.

Figure 9: GET request with C2 server

After conversion to ASCII, the next JavaScript is revealed, and the code is executed. This JavaScript comes with an embedded PE payload which may be either a loader for Gootkit, or for the REvil ransomware. There are also some differences in the algorithm used to deobfuscate it.

Example for the Gootkit variant (commented, full)

Figure 10: The downloaded JavaScript

The downloaded code chunk is responsible for installing the persistent elements. It also runs a Powershell script that reads the storage, decodes it and runs it further.

Stage 3 – The stored payload and the decoding Powershell

The authors diversified the method of encoding and storing the payload. During our tests we observed two ways of encoding. In one of them, the PE is stored as a Base64 encoded string, and in the other as a hexadecimal string, obfuscated by having certain numbers substituted by a pattern.

The payload is usually stored as a list of registry keys, yet we also observed a variant in which similar content was written into a TXT file.

Example of the payload stored in a file:

Figure 11: Payload as a file on disk

The content of the file is an obfuscated Powershell script that runs another Base64 obfuscated layer that finally decodes the .NET payload.

Example of the Powershell script that runs to deobfuscate the file:

"C:\Windows\SysWOW64\WindowsPowerShell\v1.0\powershell.exe" -ExecutionPolicy Bypass -windowstyle hidden -Command "IEX (([System.IO.File]::ReadAllText('C:\Users\[username]\bkquwxd.txt')).Replace('~',''));"

Below we will study two examples of the loader: One that leads to execution of the REvil ransomware, and another that leads to the execution of Gootkit.

Example 1—Loading REvil ransomware

The example below shows the variant in which a PE file was encoded as an obfuscated hexadecimal string. In the analyzed case, the whole flow led to execution of REvil ransomware. The sandbox analysis presenting this case is available here.

Execution of the second stage JavaScript leads to the payload being written to the registry, as a list of keys. The content is encoded as hexadecimal, and mildly obfuscated.

Figure 12: Fragment of the payload stored in the registry, encoded as a hexadecimal string obfuscated with a pattern

After writing the keys, the JavaScript deploys a PowerShell command that is responsible for decoding and running the stored content.

Figure 13: The JS component deploys PowerShell with a Base64 encoded script

Decoded content of the script:

Figure 14: Decoded content

It reads the content from the registry keys and deobfuscates it by substituting patterns. In the given example, the pattern “!@#” in the hexadecimal string was substituted by “1000”, then the PE was decoded and loaded with the help of .NET Reflection.

The next stage PE file (.NET):

The .NET loader comes with a hardcoded string that is the next stage PE: the final malicious payload. The Setup function called by the PowerShell script is responsible for decoding and running the next PE:

Figure 15: Hardcoded string (PE)
Figure 16: Deploying the payload

The loader runs to the next stage with the help of Process Hollowing – one of the classic methods of PE injection.

Figure 17: REvil ransom note

Example 2 – Loading Gootkit

In an other common variant, the payload is saved as Base64. The registry keys compose a PowerShell script in the following format:

$Command =[System.Text.Encoding]::Unicode.GetString([System.Convert]::FromBase64String("[content]")); Invoke-Expression $Command;Start-Sleep -s 22222;
Figure 18: Registry key storing payload

After decoding the base64-encoded content, we get another PowerShell script:

Figure 19: More PowerShell

It comes with yet another Base64-encoded piece that is further decompressed and loaded with the help of Reflection Assembly. It is the .NET binary, similar to the previous one.

The script calls a function “Install1” from the .NET module. This function loads another PE, that is embedded inside as a base64 encoded buffer:

Figure 20: Another buffer
Figure 21: Deploying the payload

This time the loader uses another method of PE injection, manual loading into the parent process.

The revealed payload is a Gootkit first stage binary: 60aef1b657e6c701f88fc1af6f56f93727a8f4af2d1001ddfa23e016258e333f. This PE is written in Delphi. In its resources we can find another PE (327916a876fa7541f8a1aad3c2270c2aec913bc8898273d545dc37a85ef7307f ), obfuscated by XOR with a single byte. It is further loaded by the first one.

Loader like matryoshka dolls with a side of REvil

The threat actors behind this campaign are using a very clever loader that performs a number of steps to evade detection. Given that the payload is stored within the registry under a randomly-named key, many security products will not be able to detect and remove it.

However, the biggest surprise here is to see this loader serve REvil ransomware in some instances. We were able to reproduce this flow in our lab once, but most of the time we saw Gootkit.

The REvil group has very strict rules for new members who must pass the test and verify as Russian. One thing we noticed in the REvil sample we collected is that the ransom note still points to instead of, indicating that this could be an older sample.

Banking Trojans represent a vastly different business model than ransomware. The latter has really flourished during the past few years and has earned criminals millions of dollars in part thanks to large ransom payments from high profile victims. We’ve seen banking malware (i.e. Emotet) turn into loaders for ransomware where different threat actors can specialize in what they do best. Time will tell what this return of Gootkit really means and how it might evolve.

Detection and protection

Malwarebytes prevents, detects and removes Gootkit and REvil via our different protection layers. As we collect indicators of compromise we are able to block the distribution sites so that users do not download the initial loader.

Our behavior-based anti-exploit layer also blocks the malicious loader without any signatures when the JavaScript is opened via an archiving app such as WinRar or 7-Zip.

Figure 22: Blocking on script execution

If a system is already infected with Gootkit, Malwarebytes can remediate the infection by cleaning up the registry entries where Gootkit hides:

Figure 23: Detection of payload hidden in registry

Finally, we also detect and stop the REvil (Sodinokibi) ransomware:

Figure 24: REvil ransomware blocked heuristically

Indicators of Compromise

Compromised websites downloading JavaScript loader:


3rd stage JavaScript C2s:


Variant 1 (Gootkit):

  1. NET loader [973d0318f9d9aec575db054ac9a99d96ff34121473165b10dfba60552a8beed4]
  2. Delphi PE [60aef1b657e6c701f88fc1af6f56f93727a8f4af2d1001ddfa23e016258e333f]
  3. PE stored in resources [327916a876fa7541f8a1aad3c2270c2aec913bc8898273d545dc37a85ef7307f]

Variant 2 (REvil):

  1. NET loader [0e451125eaebac5760c2f3f24cc8112345013597fb6d1b7b1c167001b17d3f9f]
  2. Delphi PE [d0e075a9346acbeca7095df2fc5e7c28909961184078e251f737f09b8ef892b6] – the ransomware
  3. PE stored in resources [a7e363887e9a7cc7f8de630b12005813cb83d6e3fc3980f735df35dccf5a1341] – a helper component

The post German users targeted with Gootkit banker or REvil ransomware appeared first on Malwarebytes Labs.

November spam roundup: Stalkers, property tips, porn, stern words and PayPal

Today we’re rounding up some of the interesting pieces of spam currently in circulation, taking in everything from housing deals to mysteriously free slices of cash. You may have seen some of these already. Hopefully we can help make up your mind about whatever’s lurking in your mailbox.

A full house of spam

Whether by accident or design, you may see spam land in your inbox reminiscent of multiple unrelated scams. It’s quite something when you don’t know if you’re looking at something ransom/blackmail related, or dating, or stolen passwords/data, or a combination of all three.

The title of the email is itself somewhat disturbing at first glance:

I am watching you every day let’s talk here [URL] I live next to you, you recognize me from the photo) after entering, I look forward to meeting you

From the getgo, we have a big stalker vibe going on. It’s a neighbour, and they’re going to…invite themselves in? What are they doing in your house? Why do they want to come in? Have they been watching you? I’ve seen many of those “I have your password and stole your files” mails that open with a line similar to “I’m watching you”. Admittedly, those claim to be watching through a webcam and not your bedroom window, but it’s still enough to set the old panic bells ringing.

Then things get very weird.

Do you like houses? Our spammer does

The tone shifts from vaguely menacing, to “rich intimate fantasies”. It’s also no longer happening in your own home, but one of several random properties close to you. If you ever wanted to meet up with a totally random stranger from the internet, in a dreamlike luxury bungalow which belongs to neither yourself or the message sender, then this is definitely the mail for you.

At this point, you may be asking yourself why you have a bunch of property tips next to what sounds like murderous dating spam. The answer is that spammers are trying to get around blocks/filters. There’s not much point spending time and effort spamming, if nobody ever sees it. If they can make use of valid services and piggyback into your mailbox, they’ll do that instead. Mail services may think twice about stopping messages coming from what are legitimate sources, even if the contents are somewhat dubious.

Skipping the security fence

There’s many ways to attempt a bypass. Splitting Bitcoin addresses and writing in languages other than English, using images, avoiding certain words or hiding the text, or piggybacking on other services. Here, they’re likely trying to take advantage of a legitimate site’s service to blast through detection. The property website in question offers the ability to send property recommendations with no need for sign up. It didn’t work for us in testing so either it only works sometimes, the site owners have switched it off, or the scammers haven’t used it at all. They’re merely imitating it to make it look as though it’s the real thing.

The spam links lead to a number of explicit sites. Whether or not you say you’re over/under 18, you’ll still be taken to graphic pornography games or adult dating websites.

A somewhat innovative method to get round spam traps, but I’m not sure what kind of success rate we’re talking about. Any process which goes from “potentially threatening”, to “houses for sale”, with a splash of “randomly taken to explicit pornography games” can’t have that big a target audience.

Users of Malwarebytes will find they’re protected from the sites linked from the initial mails, and also further clickthroughs/redirections:


The case of the unfriendly 419 spam

Another day, another attempt to part you from money. This 419 style missive takes the form of someone, er, telling you off. A lot. It reads as though you’re halfway through some shadowy, clandestine operation. Did I mention you’re being told off? Because that happens. A lot.

Some salient extracts:

Sometimes, I do wonder if you are really, really with your senses. How Could you keep trusting people and at the end you will lose your hard Earned money, or are you being deceived by their big names? They Impersonate on many offices, claiming to be Governors, Directors/Chairman of one Office or the other.

Their game plan is only just to extort your hard Earned money. Now, the question is how long you will continue to be Deceived? Sometimes, they will issue you fake check, introduce you to fake Diplomatic delivery, UN-existing online banking and they will also fake wire transfer of Your fund with Payment Stop Order and even send you fake ATM cards etc.

If this doesn’t feel like someone winning your confidence, you’d be right. It gets worse:

Anyway, by the virtue of my position I have been following this Transaction from inception and all your efforts towards realizing the Fund. More often than not, I sit down and laugh at your ignorance and That of those who claim they are assisting you, it is very unfortunate That at the end you loose.

Please I beseech you to stop pursuit of shadows and being Deceived. Feel free to contact me immediately as you receive this mail so that I can Explain to you the modus-operandi guiding the release of your Payment. Do not panic, be rest assured that this arrangement will be Guided by your Embassy here in Nigeria.

I do wonder what the success rate is for this one.

Lazy phishers and bad phishing pages

This is possibly the laziest or worst phish page I’ve ever seen. It starts off reasonably enough for a scam, claiming to be from a bank manager telling you there’s vast sums of unclaimed funds.

The main hook of the mail reads as follows:

As the regional Bank Manager of BOA BANK. It is my duty to send a financial report to my head office at the end of each year On the course of the 2019 year report, We discovered an excess profit of Eight Million Us Dollars, Which we have kept in SUSPENSE ACCOUNT without any beneficiary. As an officer of the bank I can not be directly connected to this Fund for Security Reasons, that is why I am contacting you for us to work together to get the said Fund. into your bank account for INVESTMENT in your Country The percentage Ratio is thus: 30% for you, 70% for me and my colleagues.

All you have to do to get the cash is fill out a form. The wheels almost immediately come off when you look at the bottom and see “Create your own Google form”.

When a phish goes off the rails

That doesn’t sound massively encouraging for a bank. All the same, it could be enough to grab some details from the unwary. That’s what I’d normally say, only for clicking the link and seeing this, the top entry for “Most depressing phish attempt in this or any other decade”:

Filling in an “Untitled form”, with an “Untitled question” containing precisely one option to select called “Option 1” and no text entry to go with it? Phenomenal and astounding, can’t see how that is going to work.

While it’s a spectacular bit of embarrassment for the scammers, it’s wonderful news for potential victims. Some serious miracle working will have to take place to part them from their money. We’ll take this as a win.

And finally…

Just a gentle reminder that fake mails claiming to be from PayPal are still doing the rounds. As per the older missives, the mail claims to be from a intl-paypal(dot)com address (it isn’t), and wants you to restore access to your account. The phishing site the mail linked to was already offline as we received it. It reads as follows:

Dear Customer,

Your account has just closed temporarily, because there is suspicious activity on your account. To avoid unwanted things, we took action to close your account temporarily. Immediate update and re-activate your account.

As part of this process, your old security info will be deleted and your contact email

Click the button below to finish update and active your info.

As always, follow the same process for the older spam runs: block, report, and delete.

Never a day goes by without a terrific volume of spam and phishing knocking at your doorstep. With any luck, we’ve given you a few pointers on who to turn away.

Stay safe, everyone!

The post November spam roundup: Stalkers, property tips, porn, stern words and PayPal appeared first on Malwarebytes Labs.

A week in security (November 23 – November 29)

Last week on Malwarebytes Labs, we talked with Chris Boyd about charities that track you online.

We also looked back at Zoom, and wondered whether it’s any safer months after its first vulnerability was reported. We talked about how Apple’s security is hampering the detection of potentially unwanted programs (PUPs). Lastly, we reported on Spotify resetting some user accounts after stolen or leaked credentials from a third-party were used in accessing them, and the US Senate passing the IoT cybersecurity Bill.

Other cybersecurity news

  • GoDaddy employees were reportedly socially engineered to assume control over several cryptocurrency services. (Source: KrebsOnSecurity)
  • A report from Check Point Security revealed that vishing, or “voice phishing”, is on the uptick. And usually employees who fall for such tactics are those working from home due to the pandemic. (Source: SecurityBrief)
  • Meanwhile, according to a survey by Juniper Networks, remote work has widened organizations’ attack surface, giving cybercriminals more opportunities to launch attacks against them. (Source: Entrepreneur)
  • Smart doorbells were found to be an “easy target for hackers”. Why are we not surprised? (Source: The BBC)
  • The FBI warned people to be careful after it found newly registered domains pretending to belong to the organization. (Source: Bleeping Computer)
  • Several Minecraft mods were found in the Google Play Store that are just adware apps and do nothing for you or for the game. (Source: CyberScoop)
  • Mustang Panda, a suspected hacking group from China, continues to gather intelligence about Vatican diplomacy due to the Catholic Church’s operations in China. (Source: CyberScoop)
  • According to a report, 38 percent of online video gamers have suffered from account hacking “at least once” in the past. (Source: Atlas VPN)

Stay safe, everyone!

The post A week in security (November 23 – November 29) appeared first on Malwarebytes Labs.

IoT cybersecurity bill passed by Senate

Days before taking a week-long Thanksgiving recess, the US Senate passed an almost mundane cybersecurity bill that, if approved by the President, will improve security guidelines and protocols for Internet of Things (IoT) devices purchased and owned by the Federal government.

The bill, called the Internet of Things cybersecurity Improvement Act of 2020, was actually introduced into the US House of Representatives last year. The Senate agreed to pass the legislation on November 17 under “unanimous consent,” which means that one Senator—in this case Senator Rob Portman of Ohio—asked that the bill be passed without any single objection from any of his colleagues. It does not mean the bill received unanimous votes in its favor. The procedural move is rare when passing legislation in the Senate.

Upon passage, Harley Geiger, director of public policy at cybersecurity company Rapid7, spoke highly of the bill.

“This is arguably the most significant US IoT-specific cybersecurity law to date, as well as the most significant law promoting private sector adoption of coordinated vulnerability disclosure,” Geiger wrote in a company blog post. “IoT security is widely acknowledged as a global priority, and vulnerability disclosure processes are fundamental security practices, so passage of the bill should be seen as a very positive step forward for cybersecurity and the security community.”

The bill focuses primarily on guidelines and procedures.

First, the IoT Improvement Act of 2020, if signed into law, will require the Director of the National Institute of Standards and Technology (NIST) to develop and publish “standards and guidelines for the Federal government on the appropriate use and management by agencies of Internet of Things devices.”

Those standards will apply to IoT devices owned and controlled by Federal government agencies, and they must provide guidance on secure development, identity management, patching, and configuration management.

After the NIST director publishes those guidelines, the bill will require that the Director of the Office of Management and Budget review the current information security policies and principles of Federal civilian agencies, and make sure that those policies line up with the NIST’s newer guidelines. That review will also require coordination with the director of the cybersecurity and Infrastructure Security Agency, or CISA, which until last week, was a position held by Chris Krebs.

Further, the current Federal acquisition rules for purchasing and owning IoT equipment must be updated in line with the required NIST guidelines to be published after the passage of the bill. As part of these requirements, a government agency will not be allowed to purchase IoT devices if that agency’s Chief Information Officer finds that such a device would fall short of the newly imposed rules.

Finally, the bill will require that NIST also develops guidelines for discovering and disclosing vulnerabilities in IoT devices that it owns or controls.

The IoT cybersecurity Improvement Act of 2020 marks a significant first step for the Federal government into placing security regulations on IoT devices. As we have repeatedly written aboutand spoken about—IoT security is a nascent landscape, and the lack of standardization across devices means that we are somehow both safer and more at risk to cybercriminals.

As Adam Kujawa said on our podcast about IoT cybersecurity this month, the best advantage we have for IoT security are that there are different platforms, different frameworks, and different protocols, which make it harder for any single group of cybercriminals to launch a wide-scale attack.

At the same time, though, Kujawa said that this scenario “works against us in the sense that developing security tools in order to protect these devices is just as difficult because you can’t create one solution that will necessarily work on every single device.”

The IoT cybersecurity Improvement Act of 2020 could help usher in a future where IoT device-makers can look to a single set of guidelines for their products. While the bill does not require these standards to be applied to devices purchased by general consumers, the guidance itself could still be helpful in creating agreed-upon security goals.

With unanimous consent from the Senate, there should be little reason for the president not to sign the IoT cybersecurity Improvement Act of 2020 into law.

The post IoT cybersecurity bill passed by Senate appeared first on Malwarebytes Labs.

Spotify resets some user logins after hacker database found floating online

A team of researchers working for vpnMentor has found a treasure trove in the form of an unsecured Elasticsearch database containing over 380 million records. The trove contained login credentials and other data belonging to Spotify users.

So what’s Spotify doing leaving its user data hanging around on an unsecured database? Answer: It’s not. On investigation, the team found the database didn’t actually belong to Spotify. Instead, the database was in use by a third party to defraud Spotify users.

What happened?

“The vpnMentor research team discovered the database as part of a huge web mapping project.”

After port scanning and examining weaknesses and vulnerabilities, the researchers habitually look for leaked data. This database was unsecured and unencrypted, so it was fully accessible for anyone that found it. After reviewing and confirming what they found, the team informed Spotify. Together they concluded that whoever owned the database had probably obtained the login credentials from an external site and used them on Spotify accounts.

The database builders may have used credential stuffing to verify whether the logins were valid for the Spotify service.  

The origin of the database

How this third party came into possession of, or managed to build, the database is as yet unknown. There is a possibility that it was obtained from vendors on the Dark Web. Either way, it’s clear that it would have taken them a great amount of work and/or money to amass such a huge database with verified accounts. An investment they surely would hope to earn back by defrauding Spotify users.

Trying not to gloat

It is hard not to gloat about someone’s misfortune in a case where the fraudsters’ database gets exposed. It looks as if the threat-actors should have read our blog about backdoors in elastic servers. The problem is that besides the researchers, there may have been others that found this exposed database and their intentions could have been malicious.

The content of the database

Besides the usernames and passwords for Spotify, many of the database records also contained personally identifiable information (PII) like:

  • email addresses
  • country of residence

Besides taking over a victim’s Spotify account, anyone with access to this database could use the PII to connect the data to other accounts of the victim, such as their social media profiles. The PII could also be used for spear phishing or even identity theft.

What do Spotify users need to do?

Spotify initiated an automated reset of passwords for all users affected. So if your credentials were in that database you should have received a notice about this password reset. If you didn’t receive such a notice but you want to reset your password anyway, you can follow this link and find the instructions there.

Unfortunately, and despite many users asking for MFA, Spotify has not yet enabled any kind of multi-factor authentication that we know of.

Re-used credentials

If you have used the same login credentials on other sites, which we advise against, you should change those passwords as well. Then go read our blog about why you don’t need 27 different passwords for some pointers.

Stay safe, everyone!

The post Spotify resets some user logins after hacker database found floating online appeared first on Malwarebytes Labs.

Apple security hampers detection of unwanted programs

Anyone who uses Malwarebytes software is probably familiar with the fact that, in addition to things like malware and adware, Malwarebytes detects potentially unwanted programs (PUPs). These are programs that exhibit a variety of unsavory behaviors, but that, for legal reasons, cannot be called malware.

PUP (n): a program that may include advertising, toolbars, and pop-ups that are unrelated to the software you downloaded. PUPs often come bundled with other software that you installed.

For the entire history of Malwarebytes software on iOS—the system that runs on iPhones, iPads, and iPod Touches—there have been things we would consider to be PUPs on the iOS App Store. However, due to limitations imposed by Apple, we’ve been completely unable to scan or remove PUPs from those devices (iPhones or iPads). This is simply the reality of working within Apple’s ecosystem.

On macOS, however, we’ve always been able to detect and remove PUPs. Unfortunately, we’re seeing the first signs that this is starting to change—not just for Malwarebytes, but for all security companies.

PUPs on the App Store?!

Although PUPs on Mac can be downloaded either from the App Store or the web, the question of why PUPs exist on the App Store at all is a key factor in the problem at hand. The answer is pretty simple: because Apple and Malwarebytes have different tolerance levels.

At Malwarebytes, we have a very low threshold of tolerance for PUP behaviors. We’re very aggressive in our detection of PUPs, and we have an amazing legal team that helps make that possible. It’s not always an easy stance to take, but it’s one we believe strongly in and are willing to spend resources defending.

Apple, on the other hand, is essentially in a monopoly position. It owns the hardware and the systems, and if it decides you shouldn’t run a particular program, you won’t be running that program without some significant efforts. This makes Apple far more vulnerable to lawsuits, and it has to take a more conservative approach towards PUPs.

As much as I’d like Apple to be tougher on PUPs, I understand why it can’t be as aggressive as we are.

This is not to say Apple won’t do anything about PUPs, it just needs more evidence of egregious behavior before it can act. We’ve successfully lobbied Apple in the past to get PUPs removed from the App Store, while other times we’ve been unsuccessful.

A new technology

Starting in macOS 10.15 (Catalina), Apple introduced a couple important new technologies. The first is support for system extensions. These differ from the older kernel extensions in that they are safer and easier for developers to create. Kernel extensions could fairly easily cause catastrophic crashes and other issues if a developer wrote poor kernel code.

The second technology is the EndpointSecurity framework, designed to provide support for all the things that security software used to use kernel extensions for.

These technologies are not open to everyone, however. Developers have to apply for entitlements to be allowed to use them. These entitlements are not easy to get. It took some time for us to get them here at Malwarebytes, and there are people who have a legitimate use case for these entitlements who have been rejected.

Once you have these entitlements, though, there’s a significant advantage to using system extensions in security software: once installed, and approved by the user, they are protected by macOS. This means that they become nearly impossible to remove, except by the software that installed them in the first place.

This is a really great feature for security software that may be targeted for removal by malware in order to not be detected. However, it turns out there’s a problem with this protection.

PUPs protected against removal

One of the common sub-groups of PUPs we detect are antivirus programs that show unwanted behaviors meeting certain criteria. As an example, a program that requires payment, but the antivirus engine it uses is available for free from another company, would be a likely candidate for detection.

Unfortunately, antivirus programs are also candidates for the system extension and EndpointSecurity entitlements. Anyone can apply for these entitlements, but you stand a much better chance of getting them if you are—or appear to be—a security company.

We’ve now seen a case where two different companies with a long history of making PUPs—including junk antivirus programs—have gotten these entitlements. Those programs now have a system extension, which cannot be removed by Malwarebytes or any other software.

In one case, the PUP in question is the most hated PUP by Mac IT admins and Mac tech shops everywhere, and was the subject of two separate class action lawsuits alleging fraudulent behavior.

The fallacy of Apple security

For many years, iOS has existed as a locked-down environment, incapable of being scanned for malware by any app. Antivirus software does not—and cannot—exist on iOS.

Yet iOS is not invulnerable to malware. It is unfortunately possible for an iPhone to get infected. The most famous case involves the Pegasus malware, created by NSO and used to infect journalist Jamal Khashoggi’s iPhone. Khashoggi had no way to determine that his phone was infected, and had to trust that Apple’s system was as secure as claimed. Unfortunately, this may have led to his demise.

This is a dramatic story that by no means embodies the impact of all iOS infections… but it does underscore the fact that they exist, and there’s little that anyone outside Apple can do about it. Since well-written malware shows no symptoms that the average person would be able to identify, an infected iOS device is likely to stay infected.

Apple’s new EndpointSecurity feature was touted as a more stable way for antivirus software to do its job than low-level kernel extensions. However, they are under Apple’s tight control, and this is the first concrete sign that control may push macOS in the direction of iOS.

At this point, it’s hard to say what the future of antivirus on macOS is. It’s obvious that Apple has at least some interest in supporting antivirus software, as evidenced by the creation of the EndpointSecurity framework. This is distinctly different from iOS, where such a framework does not exist.

However, it is starting to look like antivirus developers will have to play by increasingly limiting rules, and that now means not being able to protect users against certain things. Worse, Mac users will be unable to manually remove those things without contortions that the average person will find quite cumbersome.

The post Apple security hampers detection of unwanted programs appeared first on Malwarebytes Labs.

Looks like we’re stuck with Zoom: Is it any safer?

Earlier this month, Zoom’s stock price took a dive on news of two promising COVID vaccines offering over 90 percent effectiveness against the virus (a third vaccine was just announced). That’s nice. Glad to know some people think this nightmare is ending soon and we’ll all go back to the office and the classroom.

But our ability to walk into a clinic and get either of these vaccines is still months away and we’re dealing, right now, with a surge of new coronavirus infections. The reality is we’re going to be stuck with Zoom for a while longer.

Earlier in the pandemic we reported on the security risks associated with Zoom. Much of it was pretty juvenile. Think Zoombombers drawing on screen using the annotate function. On the other hand, there are countless stories online of meetings being interrupted by attendees scrawling racial epithets on screen, posting pornographic images, and threatening presenters with acts of violence. It was also revealed that Zoom’s encryption wasn’t as secure as the company claimed.

As you prepare to log in to your next Zoom meeting or class, let’s take another look at Zoom. Has it gotten any safer?


Zoom has several existing settings that users can leverage against potential meeting interlopers. That’s all well and good, but when you’re in the middle of defending your doctoral dissertation and you’re suddenly staring at a giant phallus someone scrawled over your Powerpoint (sadly, this actually happened), there’s just no good option short of shutting down your entire meeting—until now.

This month, Zoom debuted three new features that can prevent or stop disruptions like these from happening.

Suspend Participant Activities

The Suspend Participant Activities option acts like a ban hammer for presenters. Hitting this switch pauses all video, audio, chat, annotation, screen sharing, recording, and Breakout Rooms. From there, the meeting organizer can report a user and they’ll be removed from the meeting immediately.

Report users

Zoom has made it easier to report disruptive users on both the web app and the desktop client. There’s also a new setting that admins can flip that allows participants to take the initiative and report users on their own.

At-Risk Meeting Notifier

Zoom has introduced the At-Risk Meeting Notifier which scans social media posts and “other websites” for publicly shared Zoom links. If the notifier finds a meeting link online, it’ll send an automated email to the account owners and admins alerting them to the potential risk. From there, the meeting organizer can delete and reschedule the meeting with a new link.

As a quick reminder, you should require pre-registration before every meeting. Otherwise, use a random meeting ID for every meeting, instead of your Personal Meeting ID, and require a passcode to enter the meeting. And for goodness sake, disable annotation for participants if you’re delivering a presentation that in no-way requires your attendees have the ability to draw on screen.


Zoom got busted back in March for its creative definition of “end to end encryption.” As reported by The Intercept, Zoom conference data was being encrypted between the user and Zoom, meaning data was safe from someone spying on your WiFI connection. However, Zoom still had the ability to access unencrypted conference data on its end, which could be a problem if Zoom was involved in a data breach. Zoom could also be forced to hand over conference data at the request of government agencies. Fortunately, Zoom started encrypting meetings for real for both free and paid users in October.

All that being said, you have every right to remain wary given Zoom’s ambiguous language around encryption. One quick fix is to use a virtual private network (VPN) like Malwarebytes Privacy, for example. With a VPN, you’re effectively creating your own secure tunnel between yourself and Zoom. However, you’re still trusting Zoom with your data once it’s on the company’s servers.

Use something else

If this post sounds like a diss on Zoom—it’s not. This reporter happens to like Zoom. You might feel otherwise. However, switching to something else is easier said than done. Your employer or your school likely has a service agreement with Zoom. Going rogue and using the conferencing software of your choosing may not be allowed or it might not be something you can afford out of pocket. If you’re in a position where you can pick whatever web conferencing software you want, here are some important considerations:

  • Does this conferencing software feature true end-to-end encryption?
  • What options are built-in for handling meetings crashers (aka Zoombombers)?
  • Do attendees need to install the application on their computer before attending a conference?

Those are just a few of the questions you should be asking. Whatever you choose, do your due diligence, pick the right conferencing software for your needs, and keep your meetings secure.

The post Looks like we’re stuck with Zoom: Is it any safer? appeared first on Malwarebytes Labs.

Lock and Code S1Ep20: Tracking the charities that track you online with Chris Boyd

This week on Lock and Code, we discuss the top security headlines generated right here on Labs and around the Internet. In addition, we talk to Chris Boyd, lead malware intelligence analyst for Malwarebytes, about charity organizations and online ad tracking. Though many might assume that these two topics have no overlap, they absolutely do.

Ad tracking itself isn’t anything new—luxury brands used to place their advertisements specifically in newspapers that delivered to high-income zip codes, and medications for age-related illnesses broadcast commercials during daytime television, when retirees are more likely to watch.

But today’s ad tracking supercharges that match-making game with a complex, opaque machinery that can track what you do online, what websites you visit, what browser you use, and even your gender, religion, and political bias.

Tune in to hear about how charity organizations utilize online ad tracking tools—and why that could concern some users—on the latest episode of Lock and Code, with host David Ruiz.

You can also find us on the Apple iTunes storeGoogle Play Music, and Spotify, plus whatever preferred podcast platform you use.

We cover our own research on:

  • Malsmoke operators abandon exploit kits in favor of social engineering scheme
  • WebNavigator Chromium browser published by search hijackers
  • Chris Krebs, director of cybersecurity and Infrastructure Security Agency, fired by President
  • IoT forecast: Running antivirus on your smart device?

Other cybersecurity news:

  • Microsoft unveiled Pluton, a new security chip for Windows PCs that the tech giant will deliver through partnerships with Intel, AMD and Qualcomm. (Source: SecurityWeek)
  • The ransomware gang known as DarkSide has announced plans to offer a distributed storage platform for affiliates. (Source: Hot for Security)
  • Facebook fixed a critical flaw in the Facebook Messenger for Android messaging app that allowed callers to listen to other users’ surroundings. (Source: BleepingComputer)
  • A Chinese state-sponsored hacking group has infected more than 200 systems across Southeast Asia with FunnyDream. (Source: ZDNet)
  • Capcom has confirmed that hackers stole customer data and files from its internal network following a ransomware attack. (Source: TechCrunch)

Stay safe, everyone!

The post Lock and Code S1Ep20: Tracking the charities that track you online with Chris Boyd appeared first on Malwarebytes Labs.