The strongest restoration line emerges after the accidental killing in the war! Led by NVIDIA, the "AI computing power team" is preparing for a fierce attack.

Wall Street financial giant Oppenheimer recently released a research report stating that NVIDIA Corporation (NVDA.US), Broadcom Inc. (AVGO.US), Monolithic Power Systems (MPWR.US), and Marvell Technology, Inc. (MRVL.US) are still the preferred stocks in the global semiconductor sector for this investment institution. This Wall Street financial giant cited the "certainty of performance + high beta attribute" and the continuous global expansion of artificial intelligence spending as the core basis for long-term bullishness on these preferred semiconductor stocks. Several senior analysts on Wall Street recently suggested that when the global stock market enters a oversold rebound window, or when there are clear signs of easing tensions in the Middle East GEO Group Inc, chip stocks that have historically outperformed the market and have mostly been undervalued by the market are likely to become one of the most important forces leading the market rebound and valuation recovery trend. It may even be the core engine of the large rebound of the NASDAQ 100 index, the "technology stock barometer," but the premise is that international oil prices and long-term government bond yields fall simultaneously. The core logic behind this is that high beta chip stocks are most sensitive to the "GEO Group Inc easing - oil price fall - reduced pressure on interest rates" chain, so once risk appetite is restored, they often rebound first and strongest. In the broader semiconductor sector, NVIDIA Corporation and Broadcom Inc. dominate the most optimistic performance growth prospects related to "chip sectors closely related to AI computing infrastructure." This sub-sector is the most sensitive to the market rebound logic and responds most quickly and vigorously. In other words, in a scenario of "risk relief rebound," chip stocks related to AI computing power are likely to be one of the core bullish directions in the market. Jan de Bruijn, the manager of the emerging market stock fund Robeco Emerging Stars Equities, which has outperformed its peers by 96% over the past year, recently stated that chip stocks focused on high-performance and advanced processing chips for artificial intelligence provide the best risk hedge tool for the prospect of a protracted Iran war. This fund manager stated that 40% of the fund's risk exposure is primarily focused on storage chips and advanced processing chip themes, which are closely related to artificial intelligence. The chip giants associated with artificial intelligence are expected to maintain strong pricing power and fundamental expansion potential even in an economic downturn or in the face of intense fluctuations in the global financial market. As the scale of models, inference pathways, and multi-modal/agent-based Agentic AI workloads exponentially expand the consumption of computational resources, capital expenditures for technological giants are increasingly focused on the concentrated AI computing infrastructure needs driven by AI capabilities. Global investors continue to anchor the narrative of the "AI bull market," centered around NVIDIA Corporation, Alphabet Inc. Class C Google TPU clusters, AMD's new product iterations, and expectations for AI computing cluster deliveries, as one of the most certain bullish narratives in the global stock market. This also means that themes related to investments in power, liquid cooling systems, and optical interconnect supply chains closely related to AI training/inference will continue to remain among the hottest investment strategies in the stock market amidst uncertainty in the political situation of the Middle East GEO Group Inc. According to the latest analyst expectations compiled by institutions, Amazon.com, Inc. together with Alphabet Inc. Class C parent company Alphabet, Meta Platforms Inc., Oracle Corporation, and Microsoft Corporation are expected to accumulate approximately $650 billion in artificial intelligence-related capital expenditures by 2026, and some analysts believe that the overall expenditure may exceed $700 billion - indicating a year-on-year increase in AI capital expenditures could exceed 70%. It is worth noting that these five major American super-tech giants are expected to accumulate about $1.5 trillion in investments between 2023 and 2026 to build an enormous AI computing infrastructure. In contrast, these tech giants have accumulated about $600 billion in total investments during the entire historical period before 2022. In the eyes of Wall Street giants Morgan Stanley, Citigroup, Loop Capital, and Wedbush, the global investment wave in AI computing hardware-based artificial intelligence infrastructure is far from over, but is just at the beginning. Driven by an unprecedented "storm of AI inference computational demand," this round of global AI infrastructure investment wave through 2030 is expected to reach a massive scale of $3 trillion to $4 trillion. AI arms race in full swing! Oppenheimer locks in on four chip giants long term "Last week we visited several companies in the Asian semiconductor supply chain," wrote Oppenheimer analyst Rick Schafer in a report to clients. "The AI computing arms race is still in full swing, with cloud service providers having an insatiable demand for AI computing infrastructure that far exceeds supply, a situation that is expected to continue well after 2027. Supply constraints related to AI computing infrastructure span multiple dimensions, with the most prominent being advanced wafer manufacturing, advanced packaging, and high-end HBM storage systems. Delivery lead times continue to be extended. As the nearly endless demand for AI continues to absorb supply, prices of chips related to AI computing infrastructure are generally rising and are likely to be passed on to more major customers." Therefore, analyst Schafer stated that he prefers chip companies that can provide "structurally strong growth, enabling significantly outperformance throughout the cycle." Furthermore, Schafer noted that dedicated integrated circuits focused on AI (AI ASIC) are still in high demand, with the Tensor Processing Unit (TPU AI computing chip cluster) led by Alphabet Inc. Class C being the leading force. He also mentioned that to support the growth trend of large language model parameters, there are "multiple large projects" continually increasing, with market participation also on a continuous rise. Schafer stated that many of the recently announced orders may not start generating actual revenue until mid-2028. This is due to a series of challenges and issues brought about by NVIDIA Corporation's AI GPU computing rack and the AI ASIC rack connection systems dominated by Alphabet Inc. Class C, as well as other issues within data centers related to high-speed connections, including differences between traditional and new architectures. Regarding NVIDIA Corporation, Schafer stated that the number of AI server racks for the Grace Blackwell and Vera Rubin architectures this year could conservatively exceed 75,000 units. He also expected that the average selling price for Vera Rubin would be 50% or more higher than Grace Blackwell, with a conservative selling price per AI computing rack unit likely approaching $7 million. In addition, Schafer added that compared to the Grace Blackwell 200 or more advanced GB300 series, Vera Rubin has higher requirements for power management systems (up to five times that of the former), which may benefit Monolithic Power on a continuous and large scale. NVIDIA Corporation, Broadcom Inc., and Marvell Technology, Inc. are considered to be the three largest winners in the global AI computing infrastructure arms race, with NVIDIA Corporation's valuation compressed to near or below the S&P 500 index level, making it more resilient to repair compared to other chip stocks. Other conclusions from Oppenheimer's research and visits include: the shortage of server CPUs has not significantly impacted traditional server growth; cloud computing service providers and NVIDIA Corporation still prefer copper cable technology benchmarks for interconnection, but will strongly use shared package optical (CPO) in "certain necessary scenarios"; cloud giants led by Alphabet Inc. Class C are broadly adopting active copper cables. Schafer also noted a series of impacts from the shortage of storage chips, pointing out that the smartphone and PC markets have experienced the greatest negative impact. Schafer stated that while the market generally expects the overall smartphone market to "decline overall," the low-end and mid-range markets are particularly vulnerable, with smartphone shipments in China dropping by nearly 20%. Conversely, in the face of rising storage prices, the flagship smartphones under Apple Inc. (AAPL.US) have shown "more resilience". Additionally, Schafer believed that the PC market is expected to decline by 11% this year, but higher-priced AI PCs may offset some of the weakness in PC sales. AI Agent Super Wave is coming! Driving continuous and explosive expansion of AI computing demand As Agent AI/Agentic AI focused on agent-based AI workflows takes center stage in the digital world, 2026 is poised to become the year of the AI agent explosion, signaling a continuous and explosive expansion of global AI computing infrastructure demand. The explosion of AI agents such as Anthropic's Claude Cowork and OpenClaw, autonomous AI agents capable of executing tasks, in 2026 is not accidental; it essentially marks the first simultaneous convergence of "model capabilities, tool protocols, development frameworks, inference costs, and terminal context capabilities." In the AI application layer, AI agents are likely to become the dominant commercial interface, as they directly translate "intelligence" into "action," indicating that AI is moving from "answering questions" to progressing to "execution, collaboration, and completion of extremely complex multi-step tasks." Companies' urgent need to improve efficiency and reduce operational costs has significantly accelerated the widespread use of two core categories of AI application software - generative AI applications and AI intelligent agents. AI intelligent agents are likely to be the ultimate trend in AI applications for the next decade, as the emergence of AI intelligent agents signifies the evolution of artificial intelligence from an information-assistance tool to a highly intelligent productivity tool. MarketsandMarkets's latest research shows that the AI intelligent agent market is expected to reach $53 billion by 2030, indicating a high year-over-year compounded annual growth rate (CAGR) of up to 46% starting from 2025. Omdia's latest research shows that the global semiconductor industry revenue could surge by over 30% in 2026, surpassing the historic milestone of $1 trillion for the first time, with such strong growth primarily driven by the robust expansion of AI training/inference computing demand that power data center storage chips, AI GPUs/AI ASICs, and data center server CPUs. In the broad chip sector, NVIDIA Corporation and Broadcom Inc. represent the AI chip/AI computing infrastructure chain that is most sensitive, responsive, and often has the greatest resilience to market rebound logic. They possess the strongest certainty of performance, clearest AI capital expenditure main line, and significant valuation recovery potential post previous retracement. Oppenheimer continues to list NVIDIA Corporation, Broadcom Inc., Monolithic Power, and Marvell Technology as preferred semiconductor stocks, reflecting the market's view of the extreme tension in AI computing infrastructure supply and demand and structural growth as the core theme of the semiconductor sector.