Tag: 사이버보안

  • WaterPlum’s 8-Month Campaign: 30,000 Devices Infected via Fake Interviews, 7,000 Wallets Drained by North Korean Hackers

    WaterPlum’s 8-Month Campaign: 30,000 Devices Infected via Fake Interviews, 7,000 Wallets Drained by North Korean Hackers

    Key Summary

    • The North Korean hacking group WaterPlum infected at least 30,000 devices across more than 100 countries during an eight-month period from December 2025 to July 2026, according to a joint advisory issued by investigative authorities from Japan, the U.S., Australia, and Germany.
    • The attack is confirmed to have compromised more than 7,000 cryptocurrency wallets, with approximately 1.7 billion yen (about $10.71 million) in crypto assets funneled to North Korea.
    • WaterPlum used job-seeking and recruitment activity as bait. The group reportedly impersonated real AI, cryptocurrency, and non-fungible token (NFT) companies, or approached victims through legitimate recruitment and freelance platforms.

    News-summary style. Compressed coverage of new cyber threat information focused on key facts, alongside threat characteristics and implications that practitioners can immediately reference.

    Table of Contents

    “It’s an interview—please run this code.” That single line infected 30,000 devices.

    Over an eight-month period from December 2025 to July 2026, the North Korean hacking group WaterPlum infected at least 30,000 devices across more than 100 countries. This was revealed in a joint advisory issued by investigative authorities from Japan, the U.S., Australia, and Germany.

    The most shocking element is the method. WaterPlum impersonated real companies in the AI, cryptocurrency, and non-fungible token (NFT) sectors, or approached job seekers through recruitment and freelance platforms. The group then dangled a “fake interview” as bait. During the interview process, victims were instructed to install specific software or run code, allowing malware to infiltrate their systems.

    More than 7,000 cryptocurrency wallets are confirmed to have been compromised in the attack. The stolen crypto is valued at approximately 1.7 billion yen, or $10.71 million. This is not a simple case of personal data breach; real assets were systematically transferred to North Korea.

    This, in the author’s view, is the most significant point: the “everyday attack surface.” The act of looking for or offering a job—something everyone does—has itself become an attack vector. It’s not phishing emails, nor is it a vulnerability exploit. This is an extreme example of social engineering that preys on the moment when a person does not suspect another person.

    Why Four Countries’ Investigative Authorities Issued the WaterPlum Advisory Simultaneously

    The fact that not just one country, but four nations simultaneously issued an advisory demonstrates the scale of WaterPlum’s operational reach. The spread of infected devices across 100 countries means that the target was not a specific region but the global pool of job seekers.

    The joining of forces by Japan, the U.S., Australia, and Germany appears to reflect the limits of any single nation’s ability to trace funds internationally and disrupt attack infrastructure. Because cryptocurrency is inherently borderless, the money moves quickly regardless of where victims are located.

    The WaterPlum Attack Path: 3 Stages of ‘Company Impersonation → Interview → Code Execution’

    The WaterPlum attack is broadly divided into three stages.

    First, the group impersonated real AI, crypto, and NFT companies, or exploited legitimate hiring platforms to approach job seekers. Second, under the pretext of a video or technical interview, they required the installation of specific programs or the execution of code from GitHub or similar sources. Third, they planted a backdoor on the victim’s device to maintain long-term access, then drained crypto wallets.

    What stands out to practitioners is that the “technical interview” setting provided WaterPlum with the perfect excuse. Asking a developer job seeker to execute code is not inherently suspicious in itself. That is likely why this attack was able to accumulate damage over eight months.

    Stage Action Result
    1. Approach Impersonate real companies; abuse hiring platforms Gain job seeker’s trust
    2. Infiltration Coerce code/software execution during technical interview Malware infiltration
    3. Theft Long-term backdoor persistence, then wallet data exfiltration Crypto asset outflow

    Why the WaterPlum Case Is Dangerous Right Now

    North Korea’s cryptocurrency theft has long been established as a core means of foreign currency acquisition. The WaterPlum case stands apart from prior incidents on two axes: scale and international coordination.

    • Infected devices: at least 30,000
    • Countries affected: 100+
    • Wallets stolen: 7,000+
    • Amount stolen: approximately 1.7 billion yen ($10.71 million)
    • Attack period: 8 months (Dec 2025 – Jul 2026)

    Most victims likely lost wallet assets without ever realizing they had been infected. A significant time gap typically exists between the moment malware infiltrates a system and the actual theft of assets.

    What to Do Right Now

    • Before any interview, cross-verify the company’s domain and business registration through official channels.
    • If you are asked to install video interview tools or run external code, report it to your IT security team immediately.
    • Separate your crypto wallets into cold and hot wallets, and implement multi-signature (multisig).
    • Install EDR products on developer devices, and run sandbox checks on any cloned repositories from GitHub or other external sources.
    • Run all files received during a hiring process through static and dynamic analysis on services like VirusTotal before submission.

    Practical Application Points

    WaterPlum’s method differs in dimension from typical malware response because it targets “what humans do most naturally.” Relying solely on technical controls will not stop this attack. Security verification steps must be embedded directly into the business processes of hiring and interviewing.

    • Recruiters: Standardize procedures to reject or require IT team verification for any requests to install video interview tools, execute code, or grant remote access.
    • Security practitioners: Establish detection rules to monitor tokens and session information being exfiltrated from job seeker devices.
    • Job seekers: Never store wallet seed phrases digitally, and use a laptop dedicated to interviews that is separate from your work device.

    Frequently Asked Questions

    What kind of organization is WaterPlum?

    WaterPlum is a North Korean hacking group, officially named in this four-country joint advisory. Its defining characteristic is the targeting of global job seekers using recruitment activity as bait.

    How does the fake interview attack proceed?

    The group offers a fake interview by impersonating a real company, then requires the installation of specific software or the execution of code from GitHub during the technical interview. In this process, malware infiltrates the device and steals crypto wallet information.

    How can I tell if I have been infected?

    If unknown processes are running in the background after an interview, or if wallet funds move without warning, you should suspect an intrusion. The fastest verification is to check EDR logs alongside your wallet transaction history.

    How is the cryptocurrency transferred to North Korea?

    Stolen wallets are laundered through mixing services, cashed out at exchanges, or converted through P2P transactions. The funds then bypass the international financial network and flow into the North Korean regime—a pattern that has been repeatedly reported.

    The WaterPlum case reconfirms that cyberattacks are no longer just a matter of technical vulnerabilities. By infiltrating a normal business process such as hiring and interviewing, this method demonstrates that “the moment you stop suspecting people” is itself the greatest vulnerability.

    Both individuals and organizations can preemptively block a significant portion of this attack by adding a single verification step to their hiring process. The full advisory content is available in the initial Boan News report.

  • Big Three Unveil AI Security Models in Coordinated Push: Inside the Google, Anthropic, and OpenAI Cyber Race

    Big Three Unveil AI Security Models in Coordinated Push: Inside the Google, Anthropic, and OpenAI Cyber Race

    Key Summary

    • Google unveiled its cybersecurity-focused AI security model, ‘Gemini 3.8 Flash Cyber,’ and self-evaluated it as the “most capable cybersecurity model to date.”
    • Google provides early access to the model to critical defense organizations in government, healthcare, and telecommunications through the ‘Fairwind Program.’
    • Based on the article headline, Anthropic and OpenAI also appear to have released cybersecurity AI models along with safeguards and access programs, but the main body of reporting only covers Google-related content.

    Analyzing how three big tech companies are applying AI in earnest to the cybersecurity field, and the policy and practical implications of restricted, trusted defender-led early access programs

    Table of Contents

    The race to develop AI security models has accelerated all at once. In September 2026, Google unveiled its cybersecurity-specialized AI security model, ‘Gemini 3.8 Flash Cyber,’ self-praising it as “the most capable cybersecurity model ever.” Reports surfaced that Anthropic and OpenAI also released their respective AI security models and early access programs in the same breath. The fact that all three big tech companies moved on the same topic simultaneously reads not as a simple product release, but as a signal of a turning point in the trend.

    Google Gemini 3.8 Flash Cyber: Where the AI Security Model Focuses

    This AI security model, called the Gemini Cyber line, narrowed its functionality to defender-centered tasks such as threat detection, breach analysis, and code security review. Rather than grafting a general-purpose generative model onto security as-is, the decisive difference from existing product lines is that this is a specialized model tailored to the workflow. The basis for Google’s use of the phrase “capable cybersecurity model” lies in its internal benchmark scores, but unless the evaluation criteria and datasets are made public, it’s difficult for readers to accept it at face value. This was the most disappointing point for me as well. The fact that benchmarks do not equal real-world performance has been repeatedly confirmed across the industry.

    Fairwind Program: The Controlled Deployment Approach of AI Security Models

    The Fairwind Program, which Google unveiled alongside the model, is a channel that provides early access to the AI security model for critical defense organizations. Organizations directly connected to national critical infrastructure—government, healthcare, telecommunications, and the like—are the priority targets, and a select group of “trusted defenders” uses the model in a controlled environment. The key point is that the program is not immediately open to general companies or individual developers. Google has chosen controlled access under the framework of responsible deployment, but the dual nature of this choice becomes the next point of contention.

    Anthropic and OpenAI Response: Limits of Reading the Headline Alone

    Based on the article headline, Anthropic and OpenAI also appear to have simultaneously released cybersecurity AI models and safeguards. However, since the main body of reporting remained at the level of an RSS summary, the specific model names, access program structure, and target organization scope of the two companies have not been confirmed. The fact that all three companies promoted AI security models at the same time is itself evidence that industry standards are forming quickly.

    Three-Company Comparison: AI Security Model Release Status at a Glance

    Item Google Anthropic OpenAI
    Model Name Gemini 3.8 Flash Cyber Unconfirmed Unconfirmed
    Early Access Program Fairwind Program Unconfirmed Unconfirmed
    Target Organizations Government, Healthcare, Telecommunications Unconfirmed Unconfirmed
    Release Scope Controlled Early Access Unconfirmed Unconfirmed
    Emphasis Selection of Trusted Defenders Estimated Safeguard-Centric Estimated Safeguard-Centric

    Issue Analysis: The Dual Edge of Responsible AI Security Model Deployment

    Controlled early access is a double-edged sword. While the advantage of preemptively reducing the possibility of model misuse is clear, it simultaneously concentrates more technological superiority in the hands of the group with access. The argument that security tools themselves can amplify asymmetry has been raised consistently in academia and policy circles. Depending on who defines the criteria for a “trusted defender,” the market may shrink, or game rules favorable to a specific group may become entrenched.

    From a practitioner’s perspective, the notable question is whether this system advances the democratization of security. If the baseline set by big tech becomes the default for global security practices, it becomes a standard in its own right. More details on the trend can be found in the original The Hacker News article.

    Practical Application Points

    • Within one week, review the scope of work that AI security models can replace in your organization’s security operations.
    • Determine in advance whether your organization qualifies as a target institution for the Fairwind Program and check the application eligibility requirements.
    • Track when competing models from the three companies release their benchmarks and build a comparative evaluation plan.
    • Check whether your team has internal guidelines for generative AI use, and draft one if it doesn’t exist.

    Actions You Can Take Right Now

    • Add “Review of AI security model adoption” as a one-line agenda item to today’s security operations meeting.
    • Update the contact information between Google Cloud Console and your security team point of contact.
    • Bookmark the official Fairwind Program guidance page and turn on notification alerts.
    • Check Anthropic and OpenAI official channels once a week for follow-up cybersecurity model announcements.
    • Add a one-page section on “Cases of generative AI used in attacks” to your internal security training materials.

    Frequently Asked Questions

    Can general companies use Gemini 3.8 Flash Cyber right away?

    A general public release date has not yet been set. The model is structured to be provided first to critical defense organizations such as government, healthcare, and telecommunications through the Fairwind Program, and general companies must wait for follow-up announcements.

    What are the eligibility requirements for the Fairwind Program?

    According to Google, organizations with critical infrastructure defense missions are the priority target. Specific eligibility requirements and procedures should be confirmed through official channels, and a CISO-level point of contact is recommended.

    What AI security models did Anthropic and OpenAI release?

    Based on the article headline, both companies appear to have released cybersecurity models, but due to limitations in the main body of reporting, the model names and access program details have not been confirmed. There is a need to watch for follow-up announcements through official channels.

    Will the introduction of cybersecurity AI reduce the demand for security personnel?

    While simple repetitive tasks may decrease, the demand for personnel responsible for model output verification and governance is likely to increase. Reorganizing the operating system, rather than just introducing tools, is the key task.

    Reference Original

    This article was written with reference to the following original: The Hacker News — Google, Anthropic, and OpenAI Unveil Cyber AI Models, Safeguards, and Access Programs