The United States Federal Bureau of Investigation (FBI) and the Department of Justice have launched a coordinated series of high-stakes enforcement actions to dismantle a massive, highly sophisticated cryptocurrency theft and social engineering operation masterminded by North Korean threat actors.

According to federal investigators, cyber operatives tied to the Democratic People’s Republic of Korea (DPRK) have siphoned off more than $400 million worth of cryptocurrency over the past 18 months by deploying an elusive and deceptive campaign codenamed "FaceStealer." In tandem with the FBI seizing the cloud-based infrastructure that powered the hard-to-detect scheme, the Department of Justice announced the court-authorized restraint and seizure of more than $1.2 billion in cryptocurrency recently stolen by North Korean-linked actors following a major hack of the Binance cryptocurrency exchange.

FBI Says North Korean Hackers Stole Over $400 Million in Crypto Using Deepfake Video Calls

United States officials emphasize that these plundered digital assets serve a singular, dangerous purpose: directly funding Pyongyang’s illicit weapons programs.

"The North Korean regime relies on proceeds from large-scale cryptocurrency theft to fund its illicit weapons program, stealing billions of dollars in recent years," said FBI Cyber Branch Operations Director Brett Leatherman in an official statement. "North Korean cyber actors have used the FaceStealer scheme to aggressively target individuals in the cryptocurrency and Web3 sectors through deceptive social engineering tactics, resulting in the theft of more than $400 million in the last 18 months."

Federal investigators trace the FaceStealer activity to a prolific, financially motivated threat cluster tracked globally as BlueNoroff. Operating under a myriad of cryptonyms—including APT38, Stardust Chollima, and UNC1069—this cybercrime arm of the Lazarus Group has built its reputation on targeting high-value financial targets, venture capital firms, blockchain protocols, and software developers.

FBI Says North Korean Hackers Stole Over $400 Million in Crypto Using Deepfake Video Calls

The mechanics of the FaceStealer campaign showcase a disturbing evolution in cyber espionage and social engineering. The operation typically begins when attackers compromise the social media and communication accounts of previous victims across platforms like Telegram, LinkedIn, Slack, and Discord. Leveraging these trusted channels, the hackers send video meeting invitations to new targets in the cryptocurrency and Web3 industries.

When a target joins the video conference, the attack mechanism springs into motion. The threat actors record the target’s live video stream while simultaneously broadcasting an authentic, prerecorded webcam video of a previous victim to cultivate an immediate sense of trust. Furthermore, the Department of Justice notes that the hackers increasingly rely on artificial intelligence and deepfake technologies to generate hyper-realistic video streams of well-known and influential figures within the blockchain and Web3 sectors, rendering the FaceStealer deception exceptionally convincing.

Moments into the meeting, victims are typically confronted with a staged technical difficulty, such as a claim that their audio cannot be heard. They are then directed to a link designed to resolve the phantom issue. This lure, styled after "ClickFix" tactics, drops malware onto the victim’s workstation, allowing the threat actors to steal private cryptocurrency keys, drain wallets, and hijack administrative privileges to critical computer and cloud infrastructure.

FBI Says North Korean Hackers Stole Over $400 Million in Crypto Using Deepfake Video Calls

Extensive research by cybersecurity firms such as JUMPSEC and Arctic Wolf has mapped the expansion of this infrastructure. What began as a targeted phishing kit impersonating videoconferencing platforms like Zoom and Microsoft Teams has evolved into a sprawling global enterprise. Arctic Wolf noted that the vast majority of identified targets operate directly in blockchain finance or adjacent investment sectors, with chief executive officers and founders accounting for nearly half of the targeted individuals.

As the campaign matured, the attackers broadened their scope, identifying hundreds of targets across dozens of countries while incorporating generative AI tools—including ChatGPT, Google models, and specialized content generators—to craft synthetic participant videos, avatars, and custom lures like fake PDF viewer updates. The FBI has successfully seized more than 900 victim webcam video recordings utilized by the hackers to impersonate trusted contacts in follow-on attacks, underscoring the scale of the harvested digital footprint.

Beyond direct financial theft, the threat actors use access gained through these social engineering operations to facilitate software supply chain attacks. By compromising the devices and repository accounts of open-source software developers and maintainers, BlueNoroff attempts to inject malicious code into widely used libraries. Earlier this year, the FaceStealer scheme was deployed against the maintainer of the popular Axios npm package in an effort to distribute a trojanized version carrying a cross-platform remote access trojan.

FBI Says North Korean Hackers Stole Over $400 Million in Crypto Using Deepfake Video Calls

From FaceStealer to Contagious Interview Campaigns

While FaceStealer represents a primary vector for direct financial plunder, security researchers note it operates alongside parallel campaigns orchestrated by North Korean operatives, most notably the recruitment-themed operations often tracked under names like "Contagious Interview" or "ClickFake Interview."

Unlike FaceStealer’s focus on video conference deception and executive impersonation, these recruitment campaigns target job seekers and software engineers by posing as tech recruiters on professional networks like LinkedIn. Victims are enticed with attractive job opportunities, only to be asked to review a take-home coding assignment or participate in a technical assessment that contains weaponized repositories or backdoored project files.

Recent analyses from firms like Sophos and JFrog have exposed the lengths to which these threat actors go to establish operational legitimacy. In late 2026, security researchers revealed that North Korean operatives established a completely fabricated decoy company named AccuraSoftHire. Complete with a professional careers site, client testimonials, active job listings on mainstream boards, and recruiters sporting verified profile badges, the fake entity was designed to lower the guard of unsuspecting developers.

FBI Says North Korean Hackers Stole Over $400 Million in Crypto Using Deepfake Video Calls

In instances documented by Sophos, a corporate macOS device belonging to an employee was compromised after the individual cloned a public code repository presented as a job interview assignment and opened it inside a workspace editor. A weaponized workspace settings file silently replaced the integrated terminal profile, executing a multi-stage payload that dropped the BeaverTail JavaScript loader and the InvisibleFerret Python backdoor.

Joint cybersecurity advisories issued by intelligence and security agencies across the United States, Japan, Australia, and Germany estimate that these recruitment-lure campaigns have compromised tens of thousands of devices globally across dozens of countries. Cybersecurity experts continue to urge employers and technical organizations to enforce strict containment protocols, such as executing interview code assignments exclusively inside disposable virtual machines rather than on primary corporate workstations, while remaining vigilant against sophisticated social engineering that exploits professional trust.

By Basiran