NVIDIA recently announced a collaboration with Seoul National University to establish the "NVIDIA AI Technology Center (NVAITC)" in South Korea. Samsung Electronics Chairman Lee Jae-Yong also traveled to San Francisco to meet with OpenAI founder Sam Altman, discussing cooperation plans in AI and semiconductors. These two industry cues point towards further expansion of global AI infrastructure and computing power layout. On the capital front, data from The Kobeissi Letter shows that the total market capitalization of global stock markets has risen to about 137% of global GDP, nearing historical highs. The pattern of high valuations being mutually reinforced by the AI narrative has become increasingly evident. Against this backdrop, according to Onchain Lens monitoring, the largest short address related to Changxin Storage recently deposited a total of about $10 million in margin into the trading platform Hyperliquid, which supports multi-asset perpetual contracts. Although the existing short positions have recorded approximately $618,000 in floating profit, this address still chose to continue increasing its shorts, reflecting a hedge and bet against the current AI-driven high valuations correcting. This demonstrates a divergence where "industrial expansion and capital optimism are in the spotlight, while on-chain shorts begin to adjust expectations behind the scenes."

NVIDIA Bets on Korea's AI Computing Landscape

While on-chain shorts are betting on a correction through structured positions, expansion on the industry side has not stalled. Recently, NVIDIA publicly announced its cooperation with Seoul National University to build the NVIDIA AI Technology Center (NVAITC). This center is located in Korea and co-built by the global GPU leader and local top universities, categorized by the media as a strategic cooperation focusing on AI technology and the semiconductor industry. From publicly available information, NVAITC is essentially a collaborative platform for cutting-edge AI research and computing power applications, aimed at deploying NVIDIA's GPU hardware, supporting software stack, and development tool systems locally, strengthening Korea's deep binding to NVIDIA's ecosystem in AI training and inference.

From a layout perspective, NVAITC extends NVIDIA's presence in Asia, especially in Korea, from "exporting computing power and chips" to "embedding into the local innovation base." In the current context where global AI training and inference heavily rely on NVIDIA's GPU and software ecosystem, such technology centers co-built with local universities are expected to create synergy in three dimensions: first, on the computing power side, helping Korean research and industry teams gain more seamless access to NVIDIA’s systems locally; second, on the talent side, locking in the next generation of developers on its platform toolchain through joint research and coursework, and project cooperation; third, on the application side, incubating more scenarios based on NVIDIA hardware and software in fields such as semiconductors, robotics, and healthcare. Although existing public materials do not disclose specific computing power scales, investment amounts, or other quantifiable metrics for NVAITC, in an environment where the total market capitalization of the global stock market is nearing historical heights, and the optimistic narrative around AI continues to ferment, such grounded nodes contribute to NVIDIA solidifying its GPU and platform voice, providing realistic support for high valuation expectations, forming a part of the broader industrial puzzle.

Samsung Chairman's Semiconductor Deal with Altman

According to Korean media reports, Samsung Electronics Chairman Lee Jae-Yong recently traveled to San Francisco to meet with OpenAI's founder Sam Altman at OpenAI's headquarters. The official statement emphasized that both sides are discussing cooperation plans in AI and the semiconductor field, without disclosing any specific terms. As a large model developer, OpenAI has a long-term demand for high-performance computing power and stable chip supply; Samsung possesses the manufacturing capabilities for storage and logic chips, providing a foundation for supplying hardware capacity for large-scale AI workloads, leaving significant room for imagination in their cooperation.

In terms of specific directions, this contact can currently be understood from a speculative perspective: cooperation could likely focus on storage systems optimized for generative AI, dedicated acceleration chips under advanced packaging processes, and medium-to-long-term supply chain security mechanisms. There is also the possibility of joint design or deep customization in certain computing power segments, but related details remain highly uncertain. Such meetings already reflect a proactive response from major semiconductor manufacturers to align with the demand for generative AI, pushing the vertical integration trend of the industry chain from models to chips to manufacturing. Observing alongside NVIDIA's establishment of NVAITC in Korea, it can be seen that the supply sides of computing power and storage are being restructured and reconfigured. In the current environment where global stock market valuations approach historical highs, the profound binding towards real capacity and supply security is becoming one of the core supporting variables to test the sustainability of AI narratives.

Global Stock Market Value Reaches 137% of GDP, Valuation Under Pressure

The Kobeissi Letter shows that the total market capitalization of the global stock market is currently about 137% of global GDP, described as "approaching historical highs," situated in an extreme valuation range of market capitalization to GDP. Although the ratio of market capitalization to GDP is merely a rough gauge, historical experience suggests that when this ratio stays at a high level over time, it often means that the overall valuation of equity assets has been fully stretched, leading to future actual returns and risk compensation space facing compression, making it more difficult to meet the risk premiums demanded by investors.

In this round of global equity market uptrend, sectors such as AI, semiconductors, and technology have contributed significantly. NVIDIA's layouts surrounding computing power and storage have reinforced the "long-term growth" expectations at the emotional level, while indirectly elevating the overall market-to-GDP ratio. As valuations rise on a macro scale, leading tech stocks in specific segments gain higher valuation tolerances due to the AI narrative. This dual layering compels market participants to reevaluate the risks of corrections. In this environment, some institutions and traders have begun to build defensive structures through options and perpetual contracts, further utilizing on-chain tools that track position and margin fluctuations to hedge potential adjustment pressures arising from extreme valuations in a scenario where traditional assets are difficult to short directly.

Changxin Storage Large Short Increases Position on Hyperliquid

In a specific case of derivative hedging structures, according to Onchain Lens monitoring, the largest short address related to Changxin Storage chose to significantly increase its position on Hyperliquid recently. This address deposited approximately $10 million in margin into the platform in two batches. While the existing short positions remain profitable, no reduction or closing of positions was made to lock in profits, instead, the decision was to maintain and expand the shorts. Monitoring shows that this address currently has about $618,000 in floating profits, indicating that the previous bearish bets have already achieved phased gains, yet the actual operation remains focused on extending the short exposure.

From a position structure perspective, the massive short position increasing margin while maintaining the same direction in a profit state is more akin to a cautious judgment of the underlying valuation or fundamentals rather than a high-frequency trading action driven by short-term emotions. Hyperliquid, as a platform that supports multi-asset synthetic trading and perpetual contracts, allows for the ongoing tracking of such large positions and margin fluctuations via on-chain or platform data, making this choice of "continuing to bet on a correction" somewhat representative in the current high valuation environment. However, the on-chain data only reflects this participant's risk preference and structural arrangement, and the future price trajectory still depends on broader market behaviors and macro variables. Thus, whether this large short position in Changxin Storage can redeem deeper correction gains remains uncertain.

Tug of AI Dividends and Valuation Pressure

In summary, with NVIDIA’s collaboration to co-build NVAITC with Seoul National University, Samsung's Chairman Lee Jae-Yong's trip to San Francisco to discuss AI and semiconductor cooperation with Sam Altman, and Jensen Huang's public emphasis that "China is destined to produce outstanding AI technology," it can be seen that global AI infrastructure and industrial synergy are still rapidly accelerating, and the optimistic narrative has not cooled down. In contrast, The Kobeissi Letter indicates that the total market capitalization of the global stock market is about 137% of global GDP, nearing historical high ranges. According to AiCoin data, the largest short address related to Changxin Storage has recently deposited about $10 million in margin on Hyperliquid, maintaining an expansion of its shorts while already having approximately $618,000 in floating profits, reflecting that there are still participants in the high valuation environment betting on a correction with cautious positions. For the crypto and on-chain market, what is worth monitoring next is whether these types of AI and semiconductor collaborations can be transformed into substantial capacity and revenue, and whether the surrounding synthetic assets and short positions will continue to expand or start to reverse. It should be emphasized that the existing data only reveal part of the participants' risk preferences and structural arrangements, and cannot prove that on-chain shorts will necessarily dominate price trends; any directional judgment must continue to be adjusted in conjunction with the evolution of subsequent fundamentals and policy changes.

Join our community, let's discuss and become stronger together!
AiCoin exclusive Hyperliquid benefits: https://app.hyperliquid.xyz/join/AICOIN88
AiCoin exclusive Aster benefits: https://www.asterdex.com/zh-CN/referral/9C50e2
On-chain Telegram community: https://t.me/AiCoinWhaleData
On-chain community: https://www.aicoin.com/link/chat?cid=N6OVMor5g
AiCoin on-chain Twitter: https://x.com/aicoinwhaledata

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。