Despite significant gains in 2026 for many semiconductor stocks closely tied to AI computing infrastructure, analysts at Wall Street giant Bank of America believe the ongoing, powerful semiconductor bull market is far from finished. Stocks essential to this AI infrastructure building frenzy have substantial room for further growth, primarily due to the likely long-term continuation of the massive wave of AI spending driven by the sustained surge in computing power demand. In the view of Bank of America's team, AI computing infrastructure is entering a more enduring and expansive capital expenditure cycle. Around the same time, another Wall Street titan, Morgan Stanley, released a report indicating the AI computing arms race is entering a phase of systemic expansion. Demand for AI infrastructure is exhibiting a rare "inelastic" trend—meaning tech giants continue to ramp up construction of AI data centers regardless of cost curves. This "demand inelasticity" is expected to persistently strengthen U.S. economic resilience and the overall earnings growth of the S&P 500 index. Meanwhile, a recent research report from commercial banking giant JPMorgan Chase suggests that, driven by the tech stock bull market frenzy dominated by the AI computing theme, the S&P 500 could breach the seemingly distant, epic milestone of 9,000 points within the next year. This implies over 20% further upside for the benchmark index, which has already surged more than 60% since 2024. Wall Street analysts are expanding the AI infrastructure narrative from a "GPU-dominated/single-core driver" story to a "full-stack computing power revaluation" encompassing AI GPU/ASIC + CPU + HBM/DRAM/NAND memory chips + optical interconnect-led high-speed data center connectivity systems. They forecast global AI infrastructure spending will approach $3 trillion by 2028. In other words, the global popularity of AI agents, combined with record-high capital expenditures from North American tech giants, is shifting the market's AI investment focus from "single-point computing power competition around GPUs" to "AI agent-driven full-stack computing power systems." The next wave of excess alpha returns will no longer belong solely to the top leaders in the AI GPU/AI ASIC space but will systematically diffuse across the full-stack AI infrastructure layer. This includes data center high-performance CPUs, DRAM/NAND/HBM memory, AI PCBs, liquid cooling systems, data center optical interconnect systems, ABF substrates/glass substrates, and a wide range of foundry services. For the U.S. stock market, which has repeatedly hit new highs and entered a new long-term bull market trajectory, and for the MSCI global equity benchmark, the increasingly fervent "AI faith" centered on the "AI computing power investment theme" has been the core and most powerful bullish driver in recent years. As long as this wave of "AI faith" continues to burn hot and sweep through global equity markets, the bull market in U.S. and global stocks will continue to chart an exceptionally strong upward curve. AI Faith Continues to Fuel the Semiconductor Bull Wave! NVIDIA and AMD Lead the AI Infrastructure Frenzy Bank of America analysts note that from an AI data center engineering perspective, the bottleneck in AI computing infrastructure is expanding from "are there enough GPUs?" to "can system-level throughput be delivered?" GPUs/ASICs handle matrix computations, CPUs manage scheduling and Agentic AI (i.e., AI agent) workflows, leaders in HBM/DRAM/NAND memory chips address large model weights, KV cache, vector databases, inference data lakes, and data movement. Data center interconnect companies like Marvell, Arista, and Astera tackle cluster communication. Semiconductor equipment giants and the EDA segment, including TSMC, Intel, ASML, Lam Research, KLA, Cadence, and Synopsys, determine advanced process nodes, advanced packaging, and yield ramp-up. NVIDIA's latest earnings report also underscores the continued exceptional strength of underlying AI computing demand. Its quarterly revenue surged 85% year-over-year to $81.62 billion, with data center revenue nearly doubling to $75.2 billion. CEO Jensen Huang described AI data center construction as "the largest infrastructure expansion in human history." NVIDIA's performance clearly highlights that the global frenzy for building AI computing infrastructure is far from over and is expanding from AI GPU/AI ASIC to data center CPUs, high-performance network infrastructure, enterprise-grade HBM/DRAM/NAND memory, server clusters at the system level, AI super-factories, and enterprise-scale AI cloud computing systems. On Wall Street, bullish sentiment for the "global AI leader," NVIDIA, is heating up further. The average Wall Street price target suggests NVIDIA's market capitalization could exceed $7 trillion. There is also a consensus that spending on computing infrastructure for AI training/inference will reach at least the $3 trillion level by 2030. According to reports, a team led by Bank of America senior analyst Wamsi Mohan wrote in a client note, "We maintain unprecedented confidence in the continued strength of AI computing infrastructure and expect NVIDIA and AMD to remain the dominant forces in this AI infrastructure frenzy. Driving factors include: (1) Sales to frontier AI labs projected to surge 3 to 5 times year-over-year; (2) Significantly improved AI monetization prospects, exponential growth in tokens (Google token statistics show a 7x year-over-year increase), and persistent concerns and potential panic among hyperscale cloud providers about being disrupted by AI technology; (3) Continued supply tightness, with aggregate utilization of all deployed infrastructure near full capacity; and (4) Underestimated sovereign, enterprise, and industrial-level AI computing demand." "We expect the total addressable market for AI-related semiconductors to grow threefold to approximately $1.7 trillion by 2030," the bank's analysts stated. This latest forecast from Bank of America is even higher than the expectation from ASML. The CEO of the Dutch lithography giant, Christophe Fouquet, stated last week that the booming global semiconductor market will face a prolonged period of supply tightness for the foreseeable future, predicting the market could reach a staggering $1.5 trillion by 2030. Strong demand for ASML's EUV (extreme ultraviolet lithography) equipment from advanced process chipmakers like TSMC and Intel, driven by seemingly insatiable needs for AI GPU/ASIC and HBM/DRAM memory chips, has helped the Dutch semiconductor equipment maker become Europe's most valuable company. Furthermore, analysts point out that the rapid rise in semiconductor stocks is being driven by earnings, as the forward price-to-earnings ratio remains around 25.6x, largely unchanged year-to-date and significantly below the historical peak of about 30x during the dot-com bubble. Analysts added, "Multiple pieces of evidence indicate that NVIDIA and AMD, along with a host of strong-performing leaders in the AI computing supply chain, remain the dominant forces in the AI infrastructure frenzy. Therefore, we see limited evidence of speculative multiple expansion. This earnings-driven fundamental factor supports the sustainability of the current rally within the context of a massive, long-term expansion of AI computing infrastructure." Consequently, Bank of America's team stated that their most favored areas in the semiconductor market remain the AI chip leaders (led by NVIDIA, AMD, and Broadcom), memory chip leaders (led by Micron), analog chips (led by Analog Devices and Texas Instruments), semiconductor equipment (led by Lam Research and KLA), EDA electronic design automation (led by Cadence), optical interconnect (led by Marvell Technology), and certain consumer electronics leaders. Regarding the central processing unit (data center CPU) computing market, Bank of America expects several CPU-related announcements at the upcoming major Computex event in Taipei next month, including more details on the data center CPU total addressable market, which NVIDIA has forecast to be as high as approximately $200 billion. In the analog chip market, the institution added that besides the major analog and MCU leaders Analog Devices and Texas Instruments, ON Semiconductor and Microchip also have relevant exposure to AI data center infrastructure, with a particular focus on ON Semiconductor's emerging AI data center power business pipeline. "AI Faith" Regardless of Cost? AI Infrastructure Spending Frenzy Moves into an Era of "Inelastic Demand" Latest signals from other financial giants regarding the semiconductor sector are largely aligned. International investment bank UBS significantly raised its price target for Micron Technology to a new Wall Street high this week, citing continued tight supply of AI-driven HBM, DRAM, and NAND. By Tuesday's market close, driven by strong bullish forces and "memory chip super cycle" investment fervor, Micron's market capitalization also surpassed $1 trillion for the first time. With Micron's stock price soaring over 800% in the past year against the backdrop of this memory chip bull market, UBS raised its target from $535 to $1,625. UBS forecasts that under its new valuation framework, Micron's market cap could approach a staggering $1.8 trillion within the next 12 months, surpassing the market capitalizations of companies like Meta, Tesla, and Berkshire Hathaway. In the view of Morgan Stanley's analyst team, AI capital expenditures are altering the macroeconomic price transmission mechanism. Traditional macroeconomic investment frameworks suggest that rising costs for copper, electricity, memory, gas turbines, and financing would naturally suppress corporate investment scale. However, current hyperscale cloud providers and frontier AI technology companies view AI computing infrastructure as strategic-level armaments, not ordinary capital goods. Morgan Stanley predicts that by 2028, nearly $3 trillion in AI-related infrastructure investment will flow through the global economy, with over 80% of the spending still ahead. A recent Morgan Stanley research report indicates that AI computing infrastructure is becoming a primary variable in global macro pricing. From the perspective of actual AI engineering and capital expenditure, Morgan Stanley emphasizes that demand for AI computing infrastructure is exhibiting a rare "inelasticity." This means that regardless of how much financing costs/sale prices rise for AI infrastructure layers—including AI GPU/AI ASIC, data center CPUs, DRAM/NAND/HBM memory, AI PCBs, liquid cooling systems, data center optical interconnect systems, ABF substrates/glass substrates, advanced packaging, broad foundry services, gas turbines, and data center power chains—tech giants remain firmly committed to adding new or massively expanding existing AI data centers and the vast energy systems supporting them. Morgan Stanley states that the AI computing arms race has entered a phase of systemic expansion. The institution has significantly revised its 2026 capital expenditure expectations for major U.S. tech giants upward from $433 billion a year ago to $805 billion. Capital expenditures for 2027 are now expected to reach $1.1 trillion, up from a previous forecast of $950 billion. Morgan Stanley's latest expectations highlight that supply chain bottlenecks at the AI infrastructure level have expanded from "mass purchasing of GPUs/ASICs" to "striving to simultaneously solve the entire AI data center delivery chain," encompassing data center power equipment, liquid cooling, data center CPUs, DRAM/NAND/HBM, optical communication/interconnect, high-performance network interconnect, transformers, gas turbines, and more. From a macro perspective, Morgan Stanley's analyst team notes that "demand inelasticity" will strengthen U.S. economic resilience but also increase uncertainty in the bond market. AI capital expenditures directly boost business fixed investment, leading the firm to raise its 2026 U.S. GDP growth forecast from 1.8% to 2.3%, and its 2027 forecast from 2.0% to 2.6%. Simultaneously, the S&P 500 earnings growth forecast was raised from 17% to 23%. Therefore, Morgan Stanley's team emphasizes that if the AI mega-trend can both stimulate corporate investment and improve corporate productivity and profit margins, then high valuations are not purely a bubble and will be strongly supported by realized earnings.
Comments