On the same day that news broke across the internet, different lines from seemingly unrelated worlds suddenly intertwined: on one side, DeepSeek planned a large-scale financing of about 50 billion RMB, with founder Liang Wenfeng intending to contribute the highest personal stake, becoming a new focus in the tech and capital markets; on the other side, security researchers disclosed that attackers utilized 13 accounts to implant a total of 575 malicious Skills on Hugging Face and ClawHub, exposing the security vulnerabilities of large model infrastructure. On the same day, there were reports that Swiss investment firm Multi Investment completed a 480 million Swiss franc financing round, publicly incorporating blockchain and Web3 into its focus areas, alongside U.S. Senator Elizabeth Warren’s criticism of the crypto project WLFI, backed by the Trump family, accusing it of benefiting insiders while harming ordinary investors, intertwined with political turmoil. Meanwhile, in the Middle East, Iran's foreign minister announced an increase of about 20% in missile stockpile and launch capability compared to February 28, while the foreign ministry issued a strong condemnation of the U.S. Navy's seizure of oil tankers. These seemingly independent signals combining in the same time window actually outline the same underlying picture: in the current acceleration of technological competition, capital games, and geopolitical confrontation, risk no longer stems from a single point event, but from an entire set of interconnected environments.

50 billion stakes placed, DeepSeek races towards AI summit

On this rapidly operating track, DeepSeek chose to debut with a highly symbolic figure — several media outlets reported that it is planning to raise about 50 billion RMB, seen as the company's first true large-scale financing round. This round is no longer merely “life-saving funds” for a “technology laboratory,” but rather interpreted by the outside world as a grand gamble focused on accelerating commercialization: transforming the "high-performance, low-cost" technology label into a clearly visible revenue curve, directly colliding with mainstream large model manufacturers.

More dramatically, founder Liang Wenfeng plans to contribute the highest personal stake in this round of financing, and this directional message itself is a clear signal — the core decision-maker chooses to place more personal stakes on the company's future. Corresponding with this is the product timeline: DeepSeek plans to launch the V4.1 model in June this year. While specific functions and technical details have not yet been disclosed, the timeline has been firmly tied to the commercialization narrative. For DeepSeek, already viewed as the representative of "high-performance, low-cost," this financing round and the upcoming V4.1 seem more like an exam of fully turning from technical reputation to the commercial battlefield, with success or failure directly determining whether it can truly have a voice in the next AI landscape.

575 malicious Skills lurking, AI development platform sounds alarm

As funds surge into the large model track, the underlying toolchain has exposed glaring cracks. Security researchers disclosed that attackers, through 13 accounts, mass-deployed malicious content on the two major development platforms, Hugging Face and ClawHub, totaling 575 malicious Skills; this scale in itself constitutes a security event that cannot be ignored. Currently available public information only confirms that these Skills possess "malicious" attributes, but does not corroborate which types of malware they belong to, nor provide a complete list of affected downstream projects or the extent of the losses. However, for any team relying on these platforms for model invocation and plugin expansion, this seems more like a ticking time bomb yet to be defused.

This incident has undeniably put the security of platforms like Hugging Face and ClawHub, considered as “infrastructure” for the AI ecosystem, under the spotlight. The so-called "skill market," third-party components, and community-contributed code are key engines for the rapid iteration of AI startups and large model applications, yet exposed clear weak links in permission boundaries, review mechanisms, and dependency management: once malicious Skills are mistakenly integrated into the production environment, attackers could insert backdoors, steal data, or tamper with reasoning logic through seemingly ordinary extension modules. For teams highly reliant on open-source models, hosted platforms, and plugin ecosystems, this structural risk will not automatically vanish with the exposure of a single event; on the contrary, it is becoming a "security cost" that must be continuously paid on the path to the commercialization of AI in the coming years.

Swiss institution increases funding for blockchain and Web3

While the AI industry is pulling between computing power and security costs, on the other hand, traditional capital is quietly locking in the next long-term track. Swiss investment institution Multi Investment recently completed a new financing round of 480 million Swiss francs, bringing its managed asset scale to exceed 3 billion Swiss francs. In its latest external statement, fintech, deep tech, and blockchain and Web3 were explicitly identified as focus areas. This is not a "test the waters" small-scale product, but a decision that directly alters the incremental decisions of asset management, indicating that from the institutional perspective, these directions have been categorized into the underlying tech stack that must be continuously bet on in the coming years.

According to currently available information, this institution has not yet disclosed which specific public chain, category of protocols, or tokens the funds will be allocated to, but within the same time window, several traditional financial institutions globally are still choosing to explore or increase crypto-related investments, forming a resonance worth noting: even with price fluctuations, regulatory controversies, and security incidents intertwining, blockchain and Web3 have not been viewed as short-term "theme investments" in institutional asset allocation charts, but rather regarded as long-term directions, alongside fintech and deep tech. This mindset itself is one of the most important signals in the current cycle.

Trump-affiliated WLFI called out, retail investor risks magnified

Unlike institutions betting on primary markets, the perspective of U.S. politics on on-chain projects has always revolved around "who is quietly profiting, and who will ultimately foot the bill." Recently, the crypto project WLFI, linked to the Trump family, was specifically named by U.S. Senator Elizabeth Warren, accusing it of “secretly monetizing” through project design, allowing insiders to exit smoothly at high positions while ordinary investors are left exposed to high risks. In her public statement, Warren pointedly stated that WLFI "enables insiders to profit while ordinary investors suffer," categorizing it as part of the group she has long criticized: crypto asset issuance and trading arrangements marked by severe information asymmetry and hard-to-penetrate conflicts of interest. However, as of now, public materials have not provided sufficient detailed on-chain data or judicial investigative conclusions to support all specific accusations, and the incident remains primarily at the level of public opinion and political defense.

Even so, this naming is already sufficient to push WLFI's transparency and governance structure into the spotlight. How the project discloses insiders' holdings and exit paths, and whether there are pre-agreed interest arrangements, is becoming a key focus for media and market scrutiny, continuing the critique by some in the U.S. political sphere that investor protection in crypto projects is inadequate. In the same time window where AI and blockchain are viewed as long-term directions by financial institutions, the WLFI turmoil reminds retail investors: tokens with strong political symbolism do not naturally acquire regulatory immunity, but rather may amplify legal uncertainty and individual investors' tail risks under the pressure of regulation and electoral narratives.

Missile capability increased by 20%, Middle East confrontation escalates

While the domestic struggle over tokens and regulation in the U.S. has not ended, in the geopolitical arena, another tense string has been pulled tighter. Iran's foreign minister recently stated through national media that compared to the baseline level of February 28, Iran's overall missile stockpile and launch capability has increased by about 20%. The statement deliberately avoided sensitive details such as specific models, ranges, and deployment locations, instead emphasizing the "overall capability," a declaration that itself carries strong deterrent connotations, effectively presenting a “capability enhancement transcript” calibrated by time coordinates before all potential adversaries.

Almost simultaneously, a firm positioning emerged on the diplomatic front. The Iranian foreign ministry also condemned the U.S. Navy’s seizure of Iranian oil tankers through official channels, directly labeling it as "aggressive behavior," and accused this action of violating previously formed ceasefire or de-escalation arrangements, despite not fully disclosing the details of the relevant mechanisms. Announcing missile capabilities as having “increased by two-tenths,” paired with the strong language surrounding the oil tankers, sends a message that transcends the single incident itself: in a region traditionally highly interconnected with global energy prices and risk appetite, the simultaneous increase in willingness to confront and military leverage has dragged the situation from fragile de-escalation into deeper uncertainty.

AI race, security cracks, and geopolitical games intertwined

From DeepSeek planning financing of about 50 billion RMB and betting on the commercialization acceleration of June’s V4.1, to Hugging Face and ClawHub having 575 malicious Skills implanted without visible systemic remediation disclosed to the outside world, both sides of the same technological rising curve have simultaneously sprouted the two scissor points of “high-leverage expansion” and “infrastructure exposure.” On the other end, as Multi Investment completes 480 million Swiss francs in funding and lists blockchain and Web3 as focal points, the Trump family-backed WLFI, under Elizabeth Warren's accusations of “insiders profiting, ordinary investors suffering,” exemplifies how regulatory forces and profit-driven impulses are tearing at this instance. These events layer a cross-section of the technology and crypto fields characterized by "capital, leverage, and systems," all tilted towards radicalism. Zooming out, Iran's claim of an approximate 20% increase in missile capability since February 28, and characterizing the U.S. seizure of oil tankers as aggression, introduces new uncertain variables beyond an already fragile ceasefire framework, also reminding the market: abrupt changes in geopolitical and energy dimensions could amplify cyclical fluctuations in tech assets, crypto projects, and even commodities. Whether this series of multi-line tensions can be calmed will depend on DeepSeek's financing and the actual effectiveness of V4.1, the substantive improvements of AI development platforms regarding security shortfalls, how regulatory paths evolve around cases like WLFI, and whether the Middle East situation escalates or returns to a controllable range.

Join our community, let’s discuss together and become stronger!
Official Telegram community: https://t.me/aicoincn
AiCoin Chinese Twitter: https://x.com/AiCoinzh
AiCoin On-Chain: https://aicoin.com/hyperliquid
AiCoin Exclusive Hyperliquid Benefits: https://app.hyperliquid.xyz/join/AICOIN88
AiCoin Exclusive Aster Benefits: https://www.asterdex.com/zh-CN/referral/9C50e2

免责声明：本文章仅代表作者个人观点，不代表本平台的立场和观点。本文章仅供信息分享，不构成对任何人的任何投资建议。用户与作者之间的任何争议，与本平台无关。如网页中刊载的文章或图片涉及侵权，请提供相关的权利证明和身份证明发送邮件到support@aicoin.com，本平台相关工作人员将会进行核查。