"High fuel ahead! From the Xeon supply shortage to the outsourcing narrative of "Terafab" HSBC shouts out a $200 target price for Intel Corporation(INTC.US)"

With the international financial giant HSBC raising its bullish target price for the long-established American chipmaker Intel Corporation (INTC.US) by 100% to $200also the highest target price for Intel Corporation among Wall Street analystsIntel Corporation has once again become one of the most watched semiconductor stocks among global retail and institutional investors. The HSBC analyst team maintains a "buy" rating on Intel Corporation, with the core logic being that the institution expects Intel Corporation's data center server CPU business and chip fabrication business to make more positive progress, which has not yet been factored into the market valuation system. Since hitting a 52-week low, Intel Corporation's stock price has skyrocketed over 470%, with a 135% increase this year driven by the AI intelligence wave. As Agentic AI rapidly becomes popular worldwide, with cloud computing giants and AI leaders accelerating their AI infrastructure spending and global AI data center computing power expansion, HSBC has made this aggressive bullish judgment. While NVIDIA Corporation has dominated the AI GPU computing market, CPUs, advanced packaging, and wafer/semiconductor manufacturing capacity spending are becoming increasingly crucial parts of the AI computing supply chain. HSBC believes that Intel Corporation can benefit from the continuous surge in data center CPU demand brought about by AI intelligence and the global wave of AI semiconductor capacity expansion led by Musk's Terafab "super chip factory". In general, HSBC has doubled Intel Corporation's target price from $100 to $200 and maintains a "buy" rating, focusing on two core manufacturing lines: the explosion in demand for data center server CPUs and the reevaluation of the chip fabrication business. HSBC states that the expansion of AI computing infrastructure demand is transitioning from simple AI GPU/TPU training to include inference, intelligent agents, server CPUs, advanced packaging, and semiconductor manufacturing capacity. According to another Wall Street financial giant, Goldman Sachs Group, Inc., the global bull market surrounding the AI computing chain is far from over, with the market focus shifting from the long-term "program/code-driven software light asset software valuation expansion" since 2008 to the "repricing of AI computing infrastructure around a series of physical assets"; i.e., the AI super bull market is transitioning from "AI chip acquisitions frenzy" to the second phase of "massive construction of AI factories"the next round of excess alpha returns will expand beyond the AI GPU/AI ASIC fields and systemically spread to the full-stack AI computing infrastructure layer, including high-performance CPUs, DRAM/NAND/HBM storage, AI PCBs, liquid cooling systems, data center optical interconnection systems, ABF substrates/glass substrates, MLCCs, electronic fabrics, and extensive wafer outsourcing. The AI computing infrastructure boom is entering a "CPU Renaissance" moment Intel Corporation is one of the world's largest semiconductor manufacturing companies, producing processors for personal computers, data center servers, network devices, and AI workloads. The company aims to manufacture chips for other customers through Intel Corporation's chip fabrication services. Headquarters are in Santa Clara, California, with a current market value of approximately $550 billion. Although there has been a significant pullback in the past week, especially with a more than 10% drop on Tuesday, Intel Corporation remains one of the best-performing stocks in the market over the past year. At the current level of around $110, the stock has risen by about 478% from its 52-week low of $18.97, but is still down by approximately 23% from its 52-week high of $142.35. The main driver of the stock's outperformance largely stems from the expectations for improved operational efficiency and accelerated transformation following the appointment of Liwu Chen as CEO of Intel Corporation. With the launch of Anthropic's Claude Cowork and other super AI agent tools like OpenClaw that can autonomously execute tasks, the global AI intelligence wave is rapidly sweeping across the world, with AI computing architecture bottlenecks transitioning from GPU cores with matrix multiplication throughput at their core to data center CPUs with control flow, task scheduling, and memory/IO coordination at their core, leading to a severe shortage of high-performance CPUs for ultra-large-scale AI data centers. In the AI computing chain, a significant amount of workload is not only spent on token generation on GPUs but also on CPU-dominated tasks such as Python interpretation, web scraping, database retrieval, RAG index access, lexical processing, task queue scheduling, RPC/IPC communication, KV state updates, etc. This means that what determines user experience increasingly relies not on the peak computing power of a single GPU but on whether the CPU has enough cores, thread concurrency, cache levels, memory bandwidth, and PCIe/CXL/interconnect scheduling capabilities to support high-frequency tool calls and high-density task switching. If CPU cores, memory subsystems, or I/O scheduling are insufficient, GPUs, even if theoretically abundant in computing power, will suffer from utilization collapse due to data preparation, task coordination, and system waiting. Over the past two years, the AI narrative has been dominated by GPUs, with CPUs playing a secondary role in the AI arms race. However, with the rise of agent-based AI workflows like OpenClaw focusing on inference workloads, data management, task scheduling, memory access, network communication, and multi-tool calls, the market has come to realize that efficient GPU clusters require a powerful CPU as the central hub of the system. This fundamentally shifts the role of CPUs from "underestimated infrastructure" back to the center stage of chip technology, with a clear implication of a "CPU Renaissance" signaling a systematic change from training-focused AI workloads to inference and intelligent agents. GPUs will still be responsible for matrix computations and model generation, but intelligent agents require continuous execution of tool calls, code execution, retrieval, database access, task scheduling, memory management, and multiprocessing orchestration, which elevates the bottleneck status of CPUs in end-to-end throughput. Intel Corporation's recent increase in server CPU prices has important market implications, as it validates that "demand exceeds supply" rather than just "cost transfer." While consumer-level Core Ultra 200S Plus SKUs have increased by $30 to $50, the more significant price hikes are seen in the data center-grade Xeon CPUsToms Hardware lists the new suggested customer price for the data center Xeon 6980P at $13,955, up from $12,460 in 2025, representing a $1,495 increase. Additionally, Intel Corporation has not uniformly raised prices for the entire Arrow Lake family, but selectively increased prices for specific SKUs that customers are willing to buy at higher prices, essentially reflecting a shift in product structure and price weighting due to supply constraints. Intel Corporation's advanced chip fabrication capabilities are poised for a rise! Chip manufacturing and EMIB advanced packaging jointly drive the second growth curve HSBC believes that Wall Street consensus expectations are overly conservative because analysts have not fully priced in the growth trajectory of Intel Corporation's chip fabrication business. In the eyes of the HSBC analyst team, the chip fabrication business (including chip manufacturing and advanced packaging) has transitioned from being an "optional item" to a core valuation variable for Intel Corporation. For the first time, HSBC has included Intel Corporation's chip fabrication services in its segmented sum valuation model, citing the widely recognized global bottleneck in advanced processes and advanced packaging capacity, the increased engagement of external customers with Intel Corporation, and the potential appearance of design commitments from the latter half of 2026 into 2027; along with the 18A process ramp-up, internal capacity reallocation, and the expansion potential of EMIB advanced packaging, Intel Corporation may not only be a CPU supplier but also an alternative platform for the "Made in America + advanced packaging" AI chip supply chain. The latest quarterly performance outlook proves that Intel Corporation's management is highly confident in the strong growth expected from the chip fabrication business. Intel Corporation has guided second-quarter revenue to be between $13.8 billion and $14.8 billion, with non-GAAP earnings per share of about $0.20. Although this guidance implies some slowdown, the management still sees a strong demand for AI computing infrastructure and the rise of chip fabrication businesses. CEO Liwu Chen stated that the next wave of AI will increasingly rely on CPUs, wafer manufacturing capacity, and advanced packaging technology, not just GPU accelerators. HSBC believes that this opportunity is much larger than what the current market perceives. For example, the institution's senior analyst, Frank Lee, has increased his forecast for server CPU shipments in 2026 from 20% to 25%, and expects it to reach 30% in 2027. In addition, HSBC expects data center and AI-related business department revenues in 2026 to be approximately $24.1 billion, 4% higher than the Wall Street consensus. Their forecast for overall revenue in 2027 is as high as $33 billion, 20% higher than the consensus. More importantly, HSBC's confidence in Intel Corporation's chip fabrication services is growing stronger. The company has for the first time included Intel Corporation's chip fabrication services in its valuation model because the demand for alternative semiconductor manufacturing capacity outside of Asia is increasing with the ambitious "chip manufacturing reshoring" led by the Trump administration, and global constraints on advanced packaging capacities could be a significant profit driver. Intel Corporation has already secured contracts with large chip fabrication business customers such as Apple Inc. and Musk's Terafab super chip factory project, while also discussing chip fabrication and advanced packaging agreements with Alphabet Inc. Class C and NVIDIA Corporation. HSBC emphasizes Intel Corporation's growth prospects in EMIB advanced packaging technology, believing that the technology has greater revenue expansion potential for the chip manufacturing business. According to the company's sensitivity analysis, the wider adoption of EMIB alone could increase Intel Corporation's earnings per share by 23% compared to baseline assumptions by 2028. SpaceX and Tesla, Inc., led by Elon Musk, the driving force behind the "Terafab" super chip factory construction project, are taking Intel Corporation's performance growth narrative from "expanding PC and data center CPU demand" further into "American AI chip fabrication and advanced packaging platform." Intel Corporation becoming a partner in the Terafab project led by Musk indicates that super AI clients are considering bypassing traditional supply chain bottlenecks and directly restructuring chip manufacturing capabilities. Intel Corporation has secured a significant manufacturing opportunity for the next-generation 14A process at Musk's Terafab project, with details still being finalized; MarketWatch cites the opinion of the UBS Group AG analyst team that SpaceX's AI-focused business could have a capital expenditure of approximately $1.1 trillion over the next five years, with semiconductor equipment expenditures related to Terafab potentially reaching around $135 billion, significantly lifting the ceiling of the semiconductor manufacturing equipment market. If Terafab becomes a large integrated semiconductor complex combining design, manufacturing, storage, advanced packaging, and testing, Intel Corporation could potentially gain higher strategic leverage through process licensing, advanced packaging, capacity cooperation, or joint venture modes, which is why institutions like Bank of America Corp and HSBC are starting to include Terafab, advanced packaging, and external wafer business in Intel Foundry's long-term revenue curve.