Here’s what happened in crypto today

In a significant development within the cryptocurrency and artificial intelligence sectors, a coordinated advisory from the United States, Germany, Japan, and Australia has exposed a sophisticated phishing operation orchestrated by North Korean hackers, which has siphoned over $10 million through deceptive cryptocurrency job offers. Concurrently, artificial intelligence research lab Anthropic has enlisted the services of Accenture to meticulously evaluate its proposed framework for moderating the rapid advancement of AI technologies. In a separate but related matter concerning corporate governance within the digital asset space, asset manager VanEck has voiced strong criticism regarding Metaplanet’s executive compensation practices, arguing that despite recent measures to mitigate shareholder dilution, the company’s management remains inadequately aligned with investor interests.

The North Korean hacking collective, identified as WaterPlum, has been instrumental in a campaign that has infected approximately 30,000 devices and resulted in the illicit acquisition of at least $10.7 million. This operation hinges on a deceptive tactic of posing as recruiters for legitimate companies operating within the cryptocurrency and artificial intelligence industries, thereby luring unsuspecting job seekers into a malicious trap. The group, also known by the alias Contagious Interview, specifically targets software developers and IT professionals across the globe. According to the joint cybersecurity advisory issued by Japan, Germany, Australia, and the United States, the fake recruiters impersonated a range of entities, including legitimate AI firms, cryptocurrency companies, and those involved in the non-fungible token (NFT) space. Furthermore, they leveraged established recruiting services to broaden their reach and enhance their credibility.

The advisory explicitly states that the primary targets of WaterPlum’s operations were individual web designers, engineers, and specialists possessing expertise in cryptocurrency, blockchain technology, and Web3 applications. This strategic targeting underscores the group’s aim to infiltrate industries at the forefront of technological innovation. The advisory also draws a direct link between WaterPlum and North Korea’s broader strategy of embedding its IT workers within foreign companies. Japanese and US authorities have assessed that the actors associated with WaterPlum, along with some North Korean IT personnel, operate under the purview of North Korea’s Munitions Industry Department, suggesting a state-sponsored element to these cybercriminal activities.

The modus operandi of WaterPlum involved enticing potential victims through various social media platforms, online job boards, gig work platforms, and freelance marketplaces. During the purported recruitment process, unsuspecting individuals were instructed to download and execute malicious files. These files were ingeniously disguised as coding assignments, essential for demonstrating technical proficiency, or as critical fixes for video-conferencing errors, a common necessity in remote work environments. Upon execution, these malicious files would compromise the victim’s device, allowing the hackers to exfiltrate sensitive information and, crucially, gain access to cryptocurrency wallets and associated credentials. The sophistication of these attacks highlights the evolving nature of cyber threats and the persistent exploitation of economic vulnerabilities in targeted nations.

In parallel to these cybersecurity concerns, the burgeoning field of artificial intelligence is grappling with its own set of ethical and safety challenges. Anthropic, a prominent AI research laboratory, has taken a proactive step by selecting Accenture, a global professional services company, as its inaugural embedded evaluator. This strategic partnership is aimed at critically assessing Anthropic’s recently proposed framework for slowing down the pace of AI development. The proposal, articulated by Anthropic CEO Dario Amodei, was unveiled on September 12th and outlines a three-step approach designed to provide adequate time for the implementation of robust safeguards. This initiative stems from growing concerns and warnings about the potential for catastrophic harm that could arise from the unchecked and rapid advancement of AI technologies.

Amodei’s proposal advocates for a deliberate pacing of AI development, emphasizing the need to balance innovation with safety. He articulated his concerns in a public post, highlighting the accelerating nature of AI progress, largely driven by the increasing capability of AI systems to contribute to the development of subsequent generations of AI. This phenomenon, termed recursive self-improvement, poses a significant challenge, as Amodei warned that if left unmanaged, it could lead to AI systems that outpace humanity’s ability to comprehend and control them. The proposal has garnered mixed reactions from prominent figures in the tech industry. OpenAI CEO Sam Altman and SpaceX CEO Elon Musk have responded positively, acknowledging the validity of Amodei’s concerns. However, Nvidia CEO Jensen Huang has expressed skepticism, arguing that such regulatory measures are unnecessary and could stifle innovation. Despite these differing viewpoints, Anthropic’s decision to engage Accenture underscores a commitment to rigorous external review and a desire to ensure that the development of advanced AI proceeds responsibly.

In a development within the financial sector, specifically concerning companies that hold significant reserves of digital assets, asset manager VanEck has publicly criticized Metaplanet’s executive compensation structure. Metaplanet, a company that holds Bitcoin on its balance sheet, has recently implemented measures aimed at curbing shareholder dilution. However, VanEck contends that these efforts fall short of establishing a compensation framework that adequately aligns the interests of management with those of its investors.

In a report published on Friday, VanEck analyzed the executive compensation practices across the ten largest digital asset treasury companies. Within this analysis, Metaplanet’s compensation structure was categorized as “Bad,” placing it in the lowest tier of assessment. VanEck’s primary concerns stem from Metaplanet’s equity plan, which amounts to 14.7% of its fully diluted shares, and its officer exposure, which stands at 8.2%.

VanEck highlighted that Metaplanet’s officer exposure is approximately ten times higher than the average of 0.8% observed among the other nine companies examined in the report. Similarly, Metaplanet’s overall equity plan is nearly four times the peer average. To provide a comparative benchmark, VanEck cited Strategy, identified as the largest corporate Bitcoin holder. Strategy’s compensation structure received a “Good” rating, with an equity plan equivalent to 2% of its fully diluted shares and officer exposure of 0.5%. VanEck noted that Strategy’s equity reserve is fixed, and any increases to the plan necessitate a shareholder vote, a mechanism that promotes greater accountability and investor oversight. The criticism from VanEck suggests a broader trend of scrutiny regarding corporate governance and executive compensation within the rapidly evolving digital asset landscape, emphasizing the importance of transparency and alignment between management and shareholders.