Threat Research

A Brief History of The Evolution of Malware

By Val Saengphaibul | March 15, 2022

A “computer virus” is one of the few transcendent technical terms everyone understands, including children. Regardless of socioeconomic background or age, everyone has an immediate negative connotation to that term. It is usually associated with something destructive to the technology we all rely on, whether it’s a laptop, smartphone, application, or gaming system, demonstrating how ubiquitous computers and technology have become in our daily lives. Part of the reason is that we have all been exposed to the impact of viruses, such as the flu or the common cold. And like its biological counterparts, a computer virus also replicates and can be transmitted from one host to another, creating problems ranging from annoying to downright destructive.

So, in recognition of over 50 years since the first computer virus was discovered, we have decided to provide a brief historical insight into the history of computer malware from the pre-internet era to the current world of botnets, ransomware, viruses, worms, and more.

To start, here is some basic terminology:

  • A virus cannot replicate without human interaction, whether clicking a link, opening an attachment, launching an application, or downloading a file.
  • A worm does not require human interaction and can replicate on its own, tunnel deep into systems, and move between devices.
  • Malware is a generic term that encapsulates all threats—viruses, worms, botnets, ransomware, etc.—anything malicious that is software-related.

It would be impossible to cover all malware and events over the past 50 years in such a short blog post. Instead, we’ve highlighted many of the most notable examples and memorable events of the past 50 years.

The early years

1971: The first Proof of Concept

Before the internet existed, at least in the form we know today, there was the Advanced Research Projects Agency Network, or ARPANET, for short. ARPANET began in 1967 to try and connect remote computers. The first computers were connected in 1969, and one year later, the Network Control Program (NCP) was developed (the predecessor to the modern TCP/IP stack). NCP was the first network transport layer to enable data to flow from computer to computer.

In 1971, the first microprocessor was developed, the Intel 4004. It was the first commercially produced general-purpose CPU. Its size (two inches rather than 12), price ($60), and performance (comparable to much larger and more expensive processors) ushered in a new era in computing

Ironically, 1971 also saw the premier of the world’s first virus Proof of Concept, dubbed “The Creeper.” Although credited and referenced by various entities as the world’s first computer virus, the Creeper actually exhibited the behavior of a worm. Based on a concept first articulated by German mathematician John von Neumann in the 1940s, it was built at BBN (an American research and development company later acquired by Raytheon) by engineer Bob Thomas. It spread through ARPANET computers and posted the following message:

"I'm the creeper, catch me if you can!"

Like its modern worm successors, it spread via a network protocol. The intent was not for malicious or devious reasons, but to see if the “I’m the creeper, catch me if you can” message could propagate to other computers via ARPANET.

It did.

1982: The first Mac virus

Contrary to what every non-technical person says, “Macs are not susceptible to viruses,” the first computer virus found in the wild, dubbed “Elk Cloner,” was designed to target Apple II computers. It was written by a then-15-year-old, who wrote such programs to play pranks on his friends. This boot sector virus propagated whenever an infected disk was run. The virus would reside in memory and look for a clean floppy disk to infect. On the fiftieth boot, Elk Cloner would display a poem to the user:

Elk Cloner: The program with a personality

It will get on all your disks

It will infiltrate your chips

Yes, it’s Cloner!

It will stick to you like glue

It will modify RAM too

Send in the Cloner!

1986: The first PC virus

In 1986, computing was still rudimentary (slow and not internet connected). The very first iterations of the internet were relegated to governments and universities. It would be at least three years before Internet Service Providers (ISPs) started providing public access to the internet, in 1989.

Sure, Bulletin Board Systems (BBS) existed, but they required making a phone call to a direct point of presence (POP) hosted by the BBS operator. Connections to the BBS were usually limited to the local audience of the BBS because phone calls to the BBS from outside the area code were billed by the minute, making them quite expensive.

However, 1986 also saw the launch of the first PC virus, dubbed “Brain.” It changed the information security world as we know it today. It originated in Pakistan but quickly spread worldwide to Europe and North America. Ironically, the virus had replicated from machine to machine because of an anti-piracy countermeasure.

The Brain virus was developed by Amjad Farooq Alvi and Basit Farooq Alvi, two brothers from Pakistan who created a boot sector virus that loaded a warning to individuals using a pirated copy of their medical software. Of course, because there wasn’t any internet, it spread through human interaction via the copying of floppy disks. Unbeknownst to the user, the master boot record (MBR) of the victim’s machine was infected while making an illegal copy of the software and spread when the disk was inserted into the next machine. Because there was no way to know that an infected MBR virus was coming along for the ride, it spread until it became a global phenomenon.

Luckily for many, it wasn’t a destructive virus. It hid a particular sector so the machine wouldn’t boot and displayed a notification that included contact information of the Farooq Alvi brothers for remediation. They claim that they wanted affected individuals to call them to discuss how to obtain their software legally.

Within the notification, it stated:

Welcome to the Dungeon (c) 1986 Amjads (pvt) Ltd VIRUS_SHOE RECORD V9.0 Dedicated to the dynamic memories of millions of viruses who are no longer with us today - Thanks GOODNESS!!! BEWARE OF THE er..VIRUS : this program is catching program follows after these messages....$#@%$@!!

Welcome to the Dungeon © 1986 Basit & Amjads (pvt). BRAIN COMPUTER SERVICES 730 NIZAM

LBOCK ALLAMA IQBAL TOWN LAHORE-PAKISTAN PHONE: 430791,443248,280530. Beware of this VIRUS.... Contact us for vaccination...

Their ingenious plan proved to be so successful that the brothers were swamped with phone calls from all over the world.

1988: The Morris worm

The difference between a virus and a worm is that a worm doesn’t need human interaction to propagate. Over 30 years ago, the world’s first worm was born. It was called the Morris worm, after its author, Robert Morris. This worm was not malicious. It was created as a Proof of Concept to see if hands-off replication was possible.

The worm contained several firsts. It exploited vulnerabilities in various programs and services and checked to see if an existing infection was present, all behaviors of modern malware. And because Morris was worried that system administrators would quarantine the worm and ignore the infection, he programmed it for persistence. However, there was no way to stop the self-replication process, so it caused high loads on devices, rendering them inoperable, and caused a denial-of-service (DoS) in networks as it spread from machine to machine.

Mr. Morris was the first convicted under the Computer Fraud and Abuse Act. However, he later became a successful entrepreneur and was awarded tenure at the Massachusetts Institute of Technology (MIT).

1989: World’s first ransomware

In 1989, the AIDS Trojan debuted, making it the worlds’ first observed ransomware. Coincidentally, internet access was also first publicly available in 1989 for the first time through an ISP called TheWorld, out of the United States. However, ransomware did not take advantage of internet connectivity to infect and target victims until 2005.

In 1989, the human AIDS virus was highly topical and relevant worldwide, similar to today’s COVID-19-related news. The AIDS Trojan was sent via mail (yes, physical mail, not email) to AIDS researchers worldwide via 20,000 infected floppy disks. Once the disk was run, it contained a questionnaire about AIDS. But on the ninetieth reboot, it changed file names to encrypted strings and hid them from the user. The screen then displayed a demand for $189 for a yearly lease or $385 for a lifetime license, sent to a PO Box in Panama. Only bankers’ drafts, cashiers’ checks, or money orders were accepted.

The AIDS Trojan was attributed to the late Dr. Joseph Popp, who claimed that he created the ransomware to donate the funds he collected to AIDS research. However, other reports state that he was upset with the World Health Organization after being rejected by them for a job. On an interesting note, Dr. Popp did not send any of his floppy disks to researchers in the U.S.

Forensic analysis of the malware revealed that the encryption key of the malware was “Dr. Joseph Lewis Andrew Popp Jr.” Dr. Popp was arrested and charged in the U.K., but during criminal proceedings, he was declared mentally unfit and deported back to the United States.

Figure 1: The AIDS Ransomware note

1992: Michelangelo

Michelangelo was the next high-profile virus to make a significant impact. Michelangelo was a boot sector virus that targeted DOS partitions. It was written in Assembly, and like its predecessors, it was spread via floppy disks due to its targeting of the master boot record and infection of attached floppy stores, allowing it to spread during the copying and loading process.

It was named Michelangelo because it had been programmed with a time bomb—an instruction to awaken on March 6, Michelangelo's birthday. What made it infamous was that after it was discovered, there was a lot of media coverage warning users to either leave their computers off on that day or change the date on their machines to a day ahead to avoid being impacted. This was the first time a virus had received so much attention from the mainstream media, in both print and television, helping to spur anti-virus software sales worldwide.

1994-95: The warez scene and the first phishing attacks

As internet usage gained traction in the United States through the launch of services such as America Online (AOL), CompuServe, and Prodigy, so did the growth of scams and phishing. For many who grew up in AOL chatrooms, the progz (slang for programs) and warez (slang for software) scene of the mid-90s was revolutionary. Because dial-up internet access was quite expensive, being provisioned by the minute, many threat actors were interested in stealing account credentials.

New programs began being traded on illicit warez chatrooms that contained punterz (to kick people offline), phishing progz (to steal user accounts), and tools used to generate random credit cards. One of the most famous programs was AOHell, a play on the name of AOL, which contained a random account creator that used randomly created credit card accounts to open an account for free for a month.

It also contained one of the first evidences of phishing. Fake automated AOL instant message bots sent indiscriminate IMs to targets asking them to verify their account credentials, claiming there was a problem with billing or some similar issue. To continue talking to the bot, the victim would have to “verify their identity” by entering their username and password. This information was then harvested by the users of the AOHell program to use or sell for free account access and spam.

Figure 2: AOHell (Wikipedia)

Other nuances seen in AOL warez rooms were programs that promised to do specific things, but that were actually credential harvesters targeting uninformed “n00bs,” or beginner users/newbies.

Welcome to Y2K

The dawn of the new millennium saw more and more people connected globally. And at the same time, there was a dramatic increase in the volume of attacks due to the increased number of potential victims, driven by the hyper-growth of the internet.

Besides the growth of new high-tech companies (the dotcom bubble), 1999 was a year mired in fear about the “Y2K” bug. While not a virus, Y2K caused widespread panic because there was a fear that legacy computers would stop operating after December 31, 1999, due to a design flaw in the BIOS, which controls the computer's motherboard. When rebooted on January 1, 2000, the operating system would believe it was January 1, 1900, disrupting everything from gas pumps and elevators to trading floors and power plants. Ultimately, this design flaw proved to be less of a problem than thought, and most organizations and individuals escaped unscathed. But the fear of Y2K dominated the news across the world for months.

1999/2000: First botnet appears

By 2000, broadband access was starting to gain traction beyond those organizations that could afford access to a T1 line via Digital Subscriber Line (DSL) connections. Home users and organizations were now able to be online 24x7. And over the next few years, cybercriminals exploited this ubiquitous access by ushering in the era of botnets and worms.

When botnets first arrived, the term invoked thoughts of Skynet, the fictitious corporate villain from the Terminator movies. But AI was still the domain of science fiction (and, of course, 22 years later, we are still only at the beginning stages of AI). But that doesn’t mean that botnets weren’t a problem. Simply put, a botnet is a group of compromised computers under the command and control of an operator. Back then, botnets were simple. They infected and spread between machines, with most botnet malware connecting to a predetermined command and control server (C2) on Internet Relay Chat (IRC—think of old AOL chatrooms) to receive instructions.

The first observed botnet was the EarthLink Spam botnet, which made its debut in 2000. It had a simple task: to send out massive quantities of spam. The EarthLink Botnet accounted for 25% of all email spam at the time, about 1.25 billion messages in total. Because of this brazen campaign, it landed an unprecedented judgment against its operator Khan C. Smith for USD$25 million.

However, the GTbot was introduced in 1999, making it the actual first botnet. In terms of malware, it was very rudimentary. It essentially spread itself to other machines and received commands via IRC. These commands were issued by GTbot controllers, who used this network of affected devices (known as zombies) to launch distributed denial-of-service (DDoS) attacks.

The rise of the worm

Worms are still part of a threat actor’s arsenal, though not as commonplace today. As explained earlier, worms are different from viruses as they do not need human interaction to spread. And because a worm propagates on its own, it can spread widely in a short amount of time. Regardless of its intention, being infected by a worm during this time period was usually quite noticeable because it often led to a denial-of-service (typically due to a flaw). Because they consume ever-increasing operating system cycles, they eventually force an infected machine to come to a grinding halt. The resulting DoS attack can cascade across an organization as the worm spreads, disrupting an entire organization, whether that was its intention or not.

2000: I LOVE YOU 2000 Blaster Sasser

The I LOVE YOU worm opened the millennium with significant media coverage because it spread around the world at record speed. The I LOVE YOU worm was created by a Onel De Guzman, a college student in the Philippines.

The I LOVE YOU worm propagated using multiple mechanisms. First, it was sent to users via email as a malicious attachment, “LOVE-LETTER-FOR-YOU.vbs.txt.” When opened by the victim, the worm would look for the victims’ Microsoft Outlook address book and send out emails impersonating the victim and replicating itself as an attachment. This novel approach caused millions of computers to be infected in days as many people trusted the emails coming from trusted associates, including friends, family, and colleagues. This method of scraping a target’s address book and impersonating them in email is still used as part of a threat actors’ tradecraft (EMOTET).

2003: Blaster (MSBlast, lovesan)

By August 2003, many individuals and organizations were connected to the internet using a broadband connection. This gave rise to record-breaking worm and wormlike attacks.

On August 11, 2003, Blaster (also known as MSBlast and lovesan) was launched. Home users and workers at large organizations were shocked when their machines suddenly experienced the dreaded “Blue Screen of Death” (BSOD) and rebooted. What they didn’t know was that they had been disrupted by the Blaster worm.

Blaster targeted a remote procedure call (RPC) vulnerability in Microsoft Windows XP and 2003 operating systems to propagate worldwide. The worm's goal was to perform a SYN flood attack against to prevent machines from accessing updates. Luckily for Microsoft, the author made the mistake of directing Blaster to the wrong domain. The domain was nonessential as machines instead used to receive their updates.

However, due to a bug in the worm, it also caused a denial-of-service (BSOD) due to buffer overflow. Continued reboots did nothing to hamper the effort, as it just started over, shutting machines off over and over. As a result of the wide adoption of internet connectivity, this became the first global denial-of-service attack.

The intentions of the authors were revealed in an ominous message found within the malware’s binary:

I just want to say LOVE YOU SAN!!

billy gates why do you make this possible ? Stop making money

and fix your software!!

The worm resulted from the reverse engineering of a Microsoft Patch Tuesday patch (colloquially known as Exploit Wednesday). Blaster did not affect organizations that had applied the (MS03-026) patches before August 11th. This example stresses the importance of organizations patching systems as quickly as possible once an update is released. Unfortunately, to this day—18 years later—many organizations still ignore this advice.

Honorable mentions

Code Red (2001)

  • This hybrid worm looked for vulnerable web servers running Microsoft IIS. Once it found a vulnerable server, it displayed the following message:

HELLO! Welcome to! Hacked By Chinese!

  • It also launched DDoS attacks against predetermined sites.

MyDoom (2004)

  • This was the fastest spreading email worm, surpassing I LOVE YOU. MyDoom still holds the record for this feat.

The Dawn of Cybercrime

2005: Mytob/Zotob, combining worms/backdoors/botnet

Before Mytob, the world of malware was mainly limited to enthusiasts who created malware out of a desire to create mischief or sheer curiosity. However, the Mytob/Zotob variants changed everything.

Mytob essentially combined the functionality of a worm/backdoor/botnet. It is a variant of MyDoom and was created by the same coder that created the Zotob worm. Mytob infected victim machines in two ways. It either arrived via email through malicious attachments or exploited vulnerabilities in the LSASS (MS04-011) protocol or RCP-DCOM (MS04-012) and used remote code execution. It also utilized the victim’s address book to propagate itself and searched for other machines via network scans to see if they could be compromised.

Mytob was one of the first viruses to specifically block or work against anti-virus software by preventing connectivity from the victims’ machine to various update sites. This was done by redirecting all known vendor URIs to, a localhost IP. This caused all queries to public-facing websites to resolve to the machine itself, essentially going nowhere.

Mytob was very prolific for its time and was consistently listed at the top of Top 10 charts. It had so many variants with different functionalities that anti-virus companies often had the entire alphabet appended to the malware name.

The Zotob variant took remnants of the Mytob source and incorporated MS05-039, which was a buffer overflow vulnerability in Microsoft Plug and Play for Windows 2000. Zotob used this variant to scan for machines vulnerable to MS05-039 to further propagate. Mytob/Zotob variants were incredibly disruptive, taking down the operations of 100 organizations, including the NY Times. It was so disruptive that even CNN News anchor Wolf Blitzer announced that Lou Dobbs could not get on the air for his regularly scheduled programming.

The era of spyware and hijacked search results

2005: CoolWebSearch and BayRob

CoolWebSearch, commonly known as “CWS, was the first cybercrime operation to hijack search results from Google, overlaying search results with those from the threat actors themselves. This was done to steal clicks from Google. CWS was most often distributed using drive-by downloads or adware programs. It was so pervasive and hard to remove that volunteers developed programs and managed web forums to help remove CWS infections for free. CWS Shredder was one of several programs widely used by victims of CoolWebSearch to help remediate their machines.

A similar attack appeared several years later, in 2007. It used a variation of hijacking search results from eBay. It was discovered when a woman in Ohio purchased a car for several thousand dollars, and the vehicle never arrived. Authorities later determined that this car had never been listed for sale and that her machine had malware that injected fake listings on her device via the BayRob malware. In a great example of cat and mouse, the FBI and Symantec waited patiently for years for the cybercriminals to make a mistake, which culminated in their arrests in 2016.

Figure 3: Screenshot of a user asking for assistance around CWS (

From spyware to spy vs. spy and the discovery of nation-state cyber weapons

2010: Stuxnet

Ushering in the 2010s was the first discovery of nation-state malware being used to target Industrial Control Services (ICS) devices—specifically, supervisory control and data acquisition devices (SCADA). Stuxnet proved to be the first specific nation-state malware targeting critical infrastructure—in this case, industrial centrifuges (specifically nuclear), causing them to overspin and cause a meltdown. Stuxnet specifically targeted organizations in Iran but soon spread to other SCADA systems around the world. Analysis of the Stuxnet malware highlighted that it wasn’t specific to Iran and could be tailored to any organization running similar ICS devices. In 2012, a NY Times article confirmed that the United States and Israel developed Stuxnet.

2011: Regin

Regin is a highly modular Remote Access Trojan (RAT). This allowed it to be highly flexible, adapting to a targeted environment. Regin was also successful because it was very innocuous in its operation. Files that were exfiltrated were often kept in an encrypted container. But instead of being stored in multiple files, everything was held in a single file, thereby avoiding arousing the suspicions of system administrators or AV software. According to Der Spiegel, Regin was a creation of the United States NSA and designed to spy on the EU and its citizens. This information was disclosed through the infamous classified information leak provided by Edward Snowden.

2012: Flame

Flame was considered the most advanced malware ever found at the time of discovery. It had everything—a wormlike capability to spread via LAN networks, it could record and capture screenshots and audio, it could eavesdrop on and record Skype conversations, and it could turn Bluetooth workstations into listening beacons that could then exfiltrate and move files to other beacons, ultimately sending files to a predetermined C2 server. Flame primarily targeted organizations in the Middle East.

The Ushering of the Modern Era of Ransomware

2011/12: Reveton

Reveton was not the first “ransomware” of the Internet-connected age. That distinction belongs to GPCODe (2005) and others. However, Reveton was the archetype of modern ransomware, helping establish the look and feel that still exists to this day, including the ubiquitous lock screen that provides details of what happened, how to get in touch with the bad actor, how to pay the ransom, and how to decrypt files, etc.

Reveton also generated a lot of press because it had all the hallmarks of being run by a professional cybercriminal organization. It was not only professional in appearance, but also utilized templates, which was another first. Lock screens would be displayed to the user based on geolocation and present the victim with a lock screen of a local law enforcement organization along with instructions on how to make payment. Reveton used so many templates that researchers often cited releases based on its fall/winter/spring/summer templates.

Figure 4: Reveton Ransom Screen with MoneyPak instructions

Figure 5: Variation with an image of former President Obama

2013: CryptoLocker - the arrival of cryptocurrency as a payment option

CryptoLocker was the first ransomware to demand payment via Bitcoin. The price for decryption was two BTC, which in 2013 (depending on the timeframe) was somewhere between $13 and $1,100, which netted the threat actors a modest sum.

Figure 6: CryptoLocker bitcoin ransom demand

Remember, this was when cryptocurrency was still in its infancy, and getting non-technical victims to not only pay but understand how to even use cryptocurrency was a hurdle to overcome.

2013: DarkSeoul and Lazarus

Besides ransomware, 2013 also ushered in the era of sinister state-sponsored attacks. DarkSeoul, as one attack was called, targeted the Korean broadcaster SBS and banking institutions in South Korea on March 20, 2013. The malware used in this attack, Jokra, targeted a device’s master boot record (MBR) and overwrote them. Many users of internet service providers, telecoms, and ATMs were also affected as their networks were taken offline. This attack was attributed to Lazarus (North Korea), which also targeted Sony Corporation in 2014 by leaking confidential information in response to “The Interview,” a film that mocked North Korean leader Kim Jong Un. The Lazarus team was also associated with attacks against the Bank of Bangladesh in 2016. They attempted to steal $951 million but only managed to get away with $81 million due to various flags in the banking transaction chain.

Figure 7: The DarkSeoul attack screen (Image Credit GIAC)

2015: Browser Locker and fake technical support scams (BSOD)

Although technically not malware, the first technical support scams and various browser locker variants first appeared in 2015. These nuisance attacks essentially mimic ransomware by causing victims to either panic and call a support number to a threat actor acting as technical support in a different country or by simply paying cryptocurrency to “clean” their system. The scheme deployed malicious JavaScript on legitimate websites that had been compromised. This JavaScript would then render a browser inoperable, frequently displaying warnings and demands in full-screen mode (payment to unlock, selling fake remediation software or technical services, etc.). Other variations of this strategy were Blue Screen of Death (BSOD) displays, which looked convincing to the victim, including a toll-free number claiming to be Microsoft technical support, but in fact, it was a threat actor in a call center in a foreign country. The threat actor would then attempt to convince the victim to provide remote access to their device for “remediation,” after which they would take control of the victim’s machine to perform further malicious acts while charging exorbitant fees for nonexistent services.

2016: The first IoT botnet arrives

The arrival of Mirai surprised many. It was the first botnet to target IoT devices. While it primarily targeted network routers, it included other IoT devices. Mirai was mainly a DDoS botnet. and was involved in notable attacks against Brian Krebs’s website, as well as being responsible for taking down a massive segment of the internet, disrupting access and services worldwide

Unlike traditional network and end-user devices, most IoT devices are not maintained. That is, they don’t receive updates on an automated basis, as a computer or smartphone does. Instead, they are often neglected and hardly are ever updated, usually because updates require that they be flashed (meaning taken offline so the software and firmware can be overwritten entirely), which can be inconvenient or even disastrous because flashed devices can be bricked (meaning permanently rendered inoperable) if done wrong. To make matters worse, many people who connect IoT devices to the internet directly do not change their default username and password. Mirai exploited this vulnerability, allowing it to propagate without any difficulty. Mirai was so prolific at one point that even famed cryptologist Bruce Schneier thought it may have been the creation of a nation-state.

Mirai was not only a high-profile attack because it was novel but also because it was able to amass a global botnet army in such a short amount of time, allowing it to redirect internet traffic to targeted sites from infected systems around the world. This made it especially hard to defend against, as the flood of traffic came from everywhere. In fact, variants of Mirai are still alive and kicking, as noted in a recent FortiGuard Labs blog, partly because the developers eventually released its code online for other criminals to use.

2017: ShadowBrokers (NSA), WannaCry, Petya/NotPetya, and not disclosing vulnerabilities

The ShadowBrokers leak of the United States National Security Agency (NSA) was unprecedented and devastating—not just because it revealed secret malware being developed at the highest levels of the U.S. government, but also because threat actors effectively repurposed the tools and exploits that were released. These tools, codenamed “Fuzzbunch,” were an exploitation framework developed by the NSA. Part of the framework included malware known as DoublePulsar, a backdoor attack that contained the infamous “EternalBlue” exploit. EternalBlue was a zero-day exploit that the NSA kept in their arsenal that targeted Microsoft’s SMB (Server Message Block) protocol (CVE-2017-0444).

It was later used to spread the infamous WannaCry, Petya/NotPetya ransomware, with disastrous consequences. These ransomware variants were so disruptive that they caused the shutdown of manufacturing facilities worldwide. Attribution for this leak was initially blamed on Russia, but to this day, nobody has been able to attribute the ShadowBrokers hack/leak to an entity.

2017: Mining for crypto

Even though cryptocurrency-related threats were relegated initially to Ransomware or cryptocurrency wallet thefts, 2018 introduced a method never seen before. XMRig is a miner application written to mine for Monero cryptocurrency and is not malicious. It works by utilizing unused CPU cycles on a machine to help solve various mathematical problems used in cryptocurrency mining. However, cybergangs began surreptitiously installing XMRig on compromised machines and devices and then collecting and aggregating the resulting data for their own crypto profit.

Common vulnerabilities exploited by various criminal attackers utilized known exploits in Apache Struts, Oracle Weblogic, and Jenkins Servers. As a result, these attacks were relegated to organizations that used these technologies and, most important to the attackers, the powerful CPUs of the devices they ran on. These vulnerabilities were also targeted because they were remotely exploitable. Compounding matters, many of these internet-facing machines were also unlikely to be patched, either through carelessness or laziness, allowing threat actors to leverage them for profit. For some fascinating insight on how these attacks worked, please reference our blog from 2018 here.

Variants of campaigns that incorporated XMRig in their attacks also targeted mobile devices via malicious Android APKs, docker containers, and supply chain attacks targeting NPM (node package manager), to list a few. For additional information on these latest attacks on the NPM supply chain, please see our Threat Signal from November, Cryptominer and Infostealer Delivered via Hijacked Popular NPM Library.

2019: GandCrab and the emergence of Ransomware as a Service

GandCrab ushered in a new wave that has escalated the volume and virulence of attacks by providing ransomware for the masses—for a fee. GandCrab sought to do two things: distance itself from the actual attacks on organizations and generate more revenue. It perfected the business model known as Ransomware-as-a-Service (RaaS). RaaS gave the GandCrab authors the luxury of working on their code while getting others to perform the actual breaches. In this model, affiliates would do all the dirty work (reconnaissance, lateral movement, delivering the ransomware, collecting the money, etc.) while the authors stayed in the background and took a cut (estimates are that they collected between 25% and 40%) of the actual ransom

This proved to be lucrative for both parties, as the authors didn’t have to take the risk of finding and infecting targets, and affiliates didn’t have to spend time trying to develop the ransomware themselves. GandCrab also appears to have followed the Agile development process, with minor and major releases deployed whenever the authors would see fit. GandCrab later announced their retirement in June of 2019, after claiming to have netted $2 billion, most likely because they were feeling the heat of the authorities. However, the GandCrab authors would later be loosely affiliated with the Sodinokibi/REvil threat actors. REvil has also been loosely affiliated to DarkSide which is most notably credited for the infamous summer 2021 attack on Colonial Pipeline. Other notable RaaS variants that have emerged since GandCrab are BlackCat, Blackmatter, Conti, and Lockbit, to name a few.


Between 1971 and early 2000, malware was mostly relegated to mischief and attempts by virus authors to see if something they had created would work. Fast forward 20+ years, and we can see that the threat landscape has evolved from mischief to include profitable cybercrime and nation-state attacks. Likewise, the evolution of the initial term “virus” to today’s all-encompassing “malware” reflects the evolution of threats over the past 20 years. It is not coincidental that the development and changes of such attacks have coincided with the development of the hyper-connected world in which we now live.

As we embark on the next 25 years—thinking ahead for the 75th anniversary of threats—we can only assume that threats will continue to target whatever technology or trends that are trending or topical. Of course, it is much easier to forecast the next 12 months, as discussed in our recent “Predictions for 2022: Tomorrow’s Threats Will Target the Expanding Attack Surface” blog. Still, one thing remains clear, the biggest single risk will always be the human element.

Learn more about FortiGuard Labs threat research and the FortiGuard Security Subscriptions and Services portfolio.  

Learn more about the Fortinet free cybersecurity training initiative, the Fortinet NSE Training programSecurity Academy program, and Veterans program.