If the crypto community can’t combat the North’s cybercrime campaign, then regulation could increase, deterring investors and dampening innovation.
“North Korea is increasingly engaging in cyber-attacks on financial institutions, especially crypto platforms, thereby increasing its funding for nuclear weapons and harming the crypto industry’s growth.”
North Korean hackers are raiding crypto platforms and building more nuclear weapons with the proceeds. The investors they steal from are left angry and guilty that their stolen money is financing weapons that threaten mankind’s very existence. As the attacks have grown in frequency and intensity, regulators have shown their teeth; they demand that something be done. If the crypto community can’t combat the North’s cybercrime campaign, then regulation could increase, deterring investors and dampening innovation.
Join us in showcasing the cryptocurrency revolution, one newsletter at a time. Subscribe now to get daily news and market updates right to your inbox, along with our millions of other subscribers (that’s right, millions love us!) — what are you waiting for?
North Korea — A Nuclear Weapons Storepile
North Korea is shrouded in mystery. Its policies, traditions and beliefs arouse suspicion, criticism and occasionally ridicule. Even its full name – The Democratic People’s Republic of North Korea – invites mockery since it’s by far and away the least democratic country on earth. The North’s Supreme Leader, Kim Jong-un, rules with a uranium-tipped hammer. He cruelly punishes those subjects who fail to follow his mystifying rules and laws. Laws that limit men and women’s hairstyles to ten pre approved styles; outlaw denim jeans because of their capitalist associations; and send three generations of families whose members break the law to labour camps or prison. Those who escape paint a wretched picture of life under Kim. And yet, despite the danger Kim Jong-un represents and the suffering he inflicts, our leaders don’t intervene. Why is this?
Cyberwarfare Enters the Building
There’s some debate around what sparked this familial flurry of interest in cyberwarfare. Most agree, however, that two core motives underpin nearly all of the Kim family’s decisions: strengthening military might through nuclear armament, and alleviating the North’s severe economic woes. It is this second, poverty-driven point that is often overlooked.
When we combine these economic woes with Kim’s bottomless desire for more military power, we better understand why he relentlessly cyberattacks banks and crypto platforms. In a word: money. And given that cyberattacks are not only inexpensive when compared to traditional warfare, but indeed profitable, we can see that it makes perfect sense for Kim to steal from his enemies rather than attack them directly. Of course, it doesn’t hurt that cyberattacks are notoriously difficult to trace, and the North can therefore somewhat plausibly deny their involvement in them (although their denials often prompt snorts of laughter and derision, as you will see later).
We don’t know much about how Kim’s cyberwarfare teams function, or how many operatives he has, or from which countries they operate. Most of what we do know comes from leaked documents and testimony from defectors and escapees. We’re not even sure what to call Kim’s hackers. They often pick a new name to hack one or two companies, only to swap it for something else once they move on. Their previous names include Guardians of the Peace, IsOne, WhoIs, NewRomanic [sic?] Cyber Army and Hidden Cobra, to name a few. Regardless of which name they use, Kim’s hackers use the same strategies and tools over and over again, which has led many to believe that most of NK’s financially motivated cyberattacks are the work of just one team known by just one name: Lazarus Group.
The earliest cyberattack attributed to Lazarus occurred between July 4-9 2009, and was dubbed Operation Troy. Nobody used the name Lazarus at the time, but the tools and strategies the hackers used were later matched up with subsequent attacks. In preparation for the attack, Lazarus agents hijacked between 20,000 and 50,000 computers, mostly in South Korea, with which they overloaded websites and servers with traffic (called a distributed denial of service, or DDoS attack). The first wave targeted U.S. and South Korean government, media and financial websites, including those of the White House, Pentagon, NYSE and the Washington Post. The second wave attacked exclusively South Korean sites, and brought down numerous administration, security, defence and intelligence sites. The third and final wave again targeted South Korean sites, including its National Intelligence Service and several of its largest banks. The true economic cost stemming from these attacks isn’t known, but estimates suggest hundreds of millions of dollars were lost from missing revenue, as the attacks prevented many people from running their businesses.
Law Enforcement Closes In
Tracing cyberattacks to any person, group, or state is a notoriously tricky business, unless of course they own up to it, which most hackers rarely do. It wasn’t surprising then that nobody — including Kim — claimed credit for wreaking havoc across South Korea’s networks. After all, why would they? What would they gain by confessing?
It is sometimes possible to trace an attacks’ origin, however, even when hackers go to great lengths to shield their identities. Operation Troy was one such case. South Korean police, during their initial investigation, found hard evidence that Lazarus was involved in the attack. And some years later, researchers proved that Lazarus reused their Operation Troy malware for other attacks. One of these attacks began on March 20, 2013, and was christened Operation Dark Seoul.
Operation Dark Seoul continued through June. Lazarus members released compromising military information about a number of NK’s most ardent enemies: chiefly the U.S. and South Korea. The leaked data included personal information about 40,000 United States Forces Korea personnel and 100,000 South Korean citizens. When the U.S. and South Korean intelligence services accused Kim of masterminding Dark Seoul, his officials claimed the U.S. and South Korea had actually cyberattacked them first, and then denied all involvement. To simultaneously claim innocence and provocation is an unusual way to defend oneself, but it’s a normal day at the office for Kim.
The Dark Seoul attacks cemented Lazarus’ reputation as a cyber force to be reckoned with, but it wasn’t until Nov. 24, 2014, that the world really woke up to what Lazarus could do. On the day in question, Sony’s employees arrived at their office expecting the usual Monday drudgery, only to find their computer screens stuck on a pixelated red skeleton adorned with the words “Guardians of Peace” and various threats (image below). Shortly thereafter, a Reddit post emerged claiming Sony Pictures had been hacked.
Questions were raised about whether this “Guardians” moniker represented a genuinely new hacking group, or if it was simply a red herring. The latter theory was true, it turns out.
Later the same day, Sony’s confidential data materialized online. The data contained a mix of social security numbers (47,000), personal and private employee emails (millions), employee salaries, as well as dozens of unpublished scripts and films. The media got right to work on the leaked emails, and it didn’t take long for juicy discussions about Hollywood stars to turn up, including Angelina Jolie and Charlie Sheen. The quantity of leaked data was actually so vast that it took journalists weeks to finish unpacking this new treasure trove of stories.
In a surprising turn of events, the hackers quickly reached out to Sony with an unusual demand: that they not release an upcoming comedy film called The Interview. Failing to shelve the film, the hackers said, would result in even more embarrassing data cropping up online. This demand raised eyebrows, alarms and red flags at Sony and the U.S. intelligence agencies, as it didn’t seem to make sense. Why would a group of evidently talented hackers go through so much trouble to prevent one movie from being released?
Once the hackers’ demand was made public, U.S. intelligence officials formally pointed the finger at Kim and his Lazarus hackers. They also cited evidence that included technical analysis, as well as similarities between the hacking tools and malware used in Sony’s hack and other attacks attributed to Lazarus. During a later investigation, the FBI found numerous proxy IP addresses that originated from somewhere in North Korea, which officially put to bed the question of “who’s to blame” for Sony’s hack.
These original and compelling counterarguments didn’t convince anyone in the U.S. government.
Where might Kim procure such talented hackers like Park? It’s not as if the North has a bustling cyber community or an expansive pool of savvy programmers from which he can simply pluck out the finest specimens and command them to steal cash on his behalf. This riddle’s answer, according to escapee testimony, is that Kim has set up a kind of cyber scouting program, wherein educators hand-pick prodigious math students to train as cyber soldiers from a young age, possibly as young as twelve. The chosen ones withdraw from their regular schools and travel to Pyongyang, where cyber experts (possibly Lazarus members) train them in cyberwarfare and to become the next generation of Lazarus hackers.
Why Attack When We Can Rob?
Kim adjusted his cyber strategy following the Sony hack. Perhaps Kim had a Supreme lightbulb moment of sorts, in which a Glorious thought appeared in his Glorious Mind that went something along the lines of “Why are we attacking our enemies when we could just rob them blind instead?”
Lazarus launched two further attacks in May and October the same year (2015), in the Philippines and Vietnam. Unlike in Ecuador, however, its attempts to steal $1M from Tien Phong Bank in Vietnam and its probing attacks on numerous Filipino banks were thwarted. Both hacks made the news, but nobody attributed either to Kim or Lazarus right away. That happened nearly a year later, following one of Lazarus’ most audacious cyberattacks: the Bangladesh Central Bank heist.
The heist began in January 2015 when someone using the name “Rasel Ahlam” sent his CV and cover letter to various employees of Bangladesh Bank – the country’s central bank. Unfortunately for those workers who read them, the attachments contained hidden malware built by the Lazarus Group. This malware gave Lazarus near total control over the bank’s network, including its internal accounts and money transfers. Surprisingly, in an extraordinary show of restraint, the hackers didn’t use this access for over a year. Instead they planned where to send their stolen money and how they would launder it.
At 20:36 EST on Feb. 4, 2016, Lazarus began the first of thirty-five attempted transfers from the Bangladesh Bank’s New York Fed account.
Kim’s hopes of a payday and some new shiny nuclear toys were nearly disappointed right away by the unlikeliest of security measures: a lone printer on the bank’s 10th floor, which printed each and every transfer from the bank’s accounts as they happened. Lazarus hackers knew about the printer before the heist, and had hacked its software to shut it down while they emptied the bank’s accounts. However, the bank’s staff noticed the printer malfunctioning and restarted it right away. After it rebooted, the printer spat out dozens of pages that showed someone was stealing $951M from their account with the U.S. Federal Reserve. The bank’s staff reached out to the Fed at once, but they couldn’t get through to anyone who could help them. Which was all part of Lazarus’ plan.
When the bank discovered the fraudulent transfers, it was 08:45 local time, so most of its staff were at their desks and ready to help. Unfortunately, the only people who actually could help worked at the Fed office in New York, USA, where the local time was 22:00. As such, most of the Fed’s employees weren’t at the office while the heist was underway and therefore couldn’t do anything to help the bank in time. By merely exploiting the time difference between the U.S. and Bangladesh, Lazarus nearly pulled off not only the biggest cyber heist in history, but the largest theft of all time. Thankfully, however, Kim’s hopes of a billion-dollar payday were dashed by sheer fluke.
The heist-ruining blunder transpired when a Lazarus member decided to send their looted cash to an RCBC bank on Jupiter Street in Manila, the Philippines. This bank’s street address would have seemed normal and innocuous enough to Lazarus, and under microscopically different circumstances they would have made off with far more money than they actually did. But by pure chance, the Fed had recently sanctioned an Iranian shipping vessel called Jupiter, so when the first batch of stolen cash arrived at a bank bearing the same name, alarm bells started ringing. The remaining transfers to the bank on Jupiter street were reviewed and then halted by Fed employees, but they weren’t fast enough to catch the first five transactions totalling $101m.
Stealing over a cool hundred mil was a remarkable achievement for Lazarus and a great payday for Kim. It was, however, only one tenth of what Kim had hoped for. Perhaps Kim’s celebratory Soju tasted somewhat sour that evening.
A New Way Forward — Hacking Crypto
After the Bangladesh heist, Lazarus & Co appear to have realised that rather than targeting banks and financial services providers, which have fiercely robust security systems in place, they should target cryptocurrency platforms instead. At first glance, this seems like an obvious bad idea. Crypto services are supposedly more secure than TradFi banks, aren’t they?
Cryptocurrencies and platforms are built using blockchains, which are designed to be decentralized (they work without a central controller approving everything, like a bank). Apps and services built on decentralized platforms have more than a few advantages over their centralized competition: they don’t have a single point of failure (and have therefore fewer cyber vulnerabilities), they operate trustlessly (without intermediaries) and blockchain data is often public, cryptographically verified, and therefore more reliable. Unfortunately, despite the cryptocurrencies themselves being secure, the same can’t be said for centralized (CeFi) exchanges.
Many crypto exchanges are far more centralized than we like to think. They often rely on just one or two crypto wallets for many or all transactions, which exposes them to the centralization-related security vulnerabilities that also trouble the TradFi sector: namely, single points of failure that are easily attacked. It is therefore the CeFi crypto exchanges that represent a chink in the crypto space’s armour. As such, it should come as no surprise that Lazarus began to target these CeFi exchanges from April 2017 onwards.
The first crypto platform targeted by Lazarus Yapizon — a small South Korean CeFi exchange who later rebranded to Youbit. Lazarus struck Yapizon on April 22, 2017, and compromised four of the exchange’s hot wallets in which most of its users’ funds were stored. The hackers made off with a third of Yapizon’s funds: 3,816 BTC, which at the time were worth about $5M.
Lazarus then targeted another CeFi platform — Bithumb — in July. At the time, Bithumb was one of the busiest crypto exchanges in South Korea and the fourth largest globally by volume. Despite its size and decent reputation, its staff didn’t notice when large sums of crypto vanished from the company wallets. It wasn’t until users found they couldn’t withdraw their funds that Bithumb’s team jumped into action. And by action, we mean they actively buried their heads in the sand. After a two-day communications blackout, the firm admitted that hackers had socially engineered their way into one of its employee’s computers, from where they stole $7M in various cryptocurrencies and the personal details of 31,800 of its users.
Kim’s hackers didn’t sit around celebrating their string of victories; they got right back to work and struck Yapizon again using the same winning strategy from the first hack. This time, they stole 4,000 Bitcoin, or 17% of the exchange’s assets. These two breaches in quick succession left Yapizon with neither money nor options. The firm filed for bankruptcy and closed its digital doors forever just a few days before Christmas. A few weeks later, South Korean intelligence blamed Lazarus for both hacks.
Blockchain — Law Enforcement's Best Friend?
It’s worth thinking about how Lazarus members cashed out so much crypto. After all, blockchain was supposed to make money laundering harder due to the technology’s transparency, wasn’t it? If that’s true, how did Kim’s cyber crew get away with so much supposedly secure money? Don’t worry –— cryptocurrencies aren’t broken, and cashing out stolen crypto isn’t a walk in the park. Just look at the millions of dollars’ worth of crypto that remains as of yet uncashed. If thieves were able to cash it out, why wouldn’t they?
Probably because it’s more trouble than it’s worth, as is often the case when fencing stolen loot. Just ask Vincenzo Peruggia, the man who stole the Mona Lisa in 1911. The art thief hid the world’s most valuable painting in a trunk in his Paris apartment for two years and then smuggled the painting into Italy where tried to sell it to an art gallery in Florence. The gallery owner promptly turned him in to the police. Peruggia was then arrested and carted off to prison, and Leonardo’s masterpiece returned to the Louvre.
Funnily enough, selling stolen crypto can be just as hard as fencing stolen artworks, if not harder — since paintings aren’t stored on a public, immutable ledger. And it’s not like Kim or a Lazarus member could easily register an account on Coinbase or Binance as both have stringent, mandatory KYC checks in place. Most of the other high-liquidity exchanges do as well.
There are, however, a few strategies and tools that Lazarus (and other hackers) use to cash out their stolen crypto. The current favourite among these are currency mixers, which let thieves conceal where their stolen crypto came from by blending it with other peoples’ crypto in a shared pool.
You may remember that during September of 2020, the Covid-19 pandemic was in full swing. Borders were shut, planes were grounded and the world economy grinded to a halt. Every country suffered from financial pain; but owing to widespread poverty, poor infrastructure, and a near non-existent list of foreign allies, North Koreans suffered worse than most. Because while other countries collaborated to create Covid tests and vaccines, Kim showed little interest in working with the West, even temporarily. Instead, he tried to steal his way to salvation.
–
So far in 2022, Lazarus has hacked two large platforms. The first was Ronin, an Ethereum sidechain built for Axie Infinity, the P2E game. The second was Harmony, a U.S.-based blockchain company that builds cross-chain bridges. .
It was six days before Ronin’s engineers realised that they had been hacked. They weren’t tipped off by any cyber tools or security measures either; it was only after a user queried why they couldn’t withdraw their Ether that the team discovered the breach. When the dust settled, 173,600 Ether and $25.5M USDC (totalling $625M) was drained from Ronin by Lazarus, making this Lazarus’ biggest score to date.
–
Once these warnings had made the rounds on Twitter, Harmony released a statement: “This incident is a humbling and unfortunate reminder of how our work is paramount to the future of this space, and how much of our work remains ahead of us.”
Can Any of This Be Stopped?
Lazarus’ relentless cyberattacks brought to the surface a question with which the crypto community has always struggled: how much control should lawmakers and regulators have over crypto in order to keep investors safe? The libertarian wing argues that no agency should be able censor, suppress or control crypto transactions, regardless of their legality; and that any control ceded to regulators is a failure. Whereas states and their regulators claim, for the most part, that they should be able to freeze any user’s funds if they suspect them of a crime, and that regulation would be the best way to constrain Kim’s cyberwarfare campaign.
There’s no easy answer to this question of how (or even if) crypto should be regulated to fight back against Lazarus. If regulators control who can and can’t use crypto, doesn’t that defeat the whole point of decentralized currencies? Equally though, if regulators could freeze criminal wallets and stamp out currency mixers, we could better fight back against Lazarus and hamper Kim’s nuclear program. Is regulation, therefore, a price worth paying?
After all, despite all its faults and weaknesses, North Korea represents one of the greatest threats to humanity’s survival. It’s also the only nation on earth whose military hacks businesses and banks to build nuclear weapons. And every time another crypto platform gets hacked, Kim’s rocket scientists produce yet another even bigger weapon of mass destruction.
Unless the crypto industry can devise a way to either recoup the money stolen during Lazarus’ hacks or somehow prevent platforms from being hacked in the first place (without help from regulators), then regulators are bound to step in soon. In a way, they already have by sanctioning Tornado Cash. If this trend spreads, if the crypto space becomes so heavily regulated that wallets and transactions require identity documents to work, then all crypto will have amounted to is a slightly more efficient version of the banking systems we have now.
What crypto needs is a concerted, global and collaborative push to develop tools and strategies that can do what regulators claim only regulation can: protect investors and prevent Lazarus from stealing any more money. If these tools don’t materialize sometime soon, regulators could use Kim as their excuse to impose KYC checks and all other kinds of regulation on the entire crypto space. Those platforms who don’t comply could face sanctions or end up blacklisted, just like Tornado Cash.