Morgan Stanley: AI chip fever is cooling down, cloud giants may see a rotation.
Related
Morgan Stanley: Funds are rotating from chip stocks to AI cloud service providers; US stocks may struggle to reach new highs in the short term.
According to Mars Finance, Morgan Stanley strategists believe that US stocks may struggle to reach new all-time highs in the short term as investors take profits on tech stocks and rotate their assets. Strategist Wilson is bullish on hyperscale cloud service providers including Microsoft, Amazon, and Meta, believing their core businesses can support volatility in AI-related businesses. He predicts the S&P 500 will reach 8,000 points by the end of the year, with consumer discretionary, transportation, and biotechnology sectors potentially benefiting. (Cailian Press)
Morgan Stanley and UBS have diverged on their AI investment themes: the former favors a rotation among cloud service providers, while the latter is betting on a revaluation of AI infrastructure.
According to Mars Finance, on July 6th, as rotation within the US AI sector intensified, Morgan Stanley and UBS offered differing assessments of the next phase of AI investment. Morgan Stanley believes that funds are shifting from the previously surging semiconductor sector to hyperscale cloud service providers such as Microsoft, Amazon, and Meta. The AI rally is not over, but rather entering a sector rotation phase. UBS, on the other hand, is more optimistic about the long-term value creation capabilities of AI infrastructure. Their Holt team predicts that the profitability of memory chip companies such as Samsung Electronics, SK Hynix, and Micron will continue to improve, and the economic profit of the AI infrastructure sector is expected to increase from approximately $200 billion in 2023 to $1.4 trillion in 2027, an increase of approximately 600%, while the economic profit of large cloud service providers is expected to only increase to approximately $400 billion during the same period. UBS believes that the memory chip industry is transforming from a traditional cyclical industry into one of the most important value creators in the AI industry chain.
Video | Morgan Stanley's Chief U.S. Equity Analyst: Chip Stocks May Correct by Over 30%, But S&P Still Targets 8000 Points by Year-End
On July 6th, Morgan Stanley's chief U.S. equity analyst, Mike Wilson, stated in an interview with Bloomberg that while chip stocks remain core assets in the AI wave, some are severely overbought, and a short-term correction of 30% to 40% is entirely possible. However, he believes this does not signify the end of the U.S. stock market bull run, and the S&P 500 is expected to continue rising as funds rotate into other sectors. He maintains his prediction that the S&P 500 will break through 8000 points by the end of the year, believing that the core logic supporting the market still comes from corporate earnings growth.
JPMorgan Chase: Semiconductor stocks' continued outperformance of cloud service providers may be unsustainable; AI trading may see sector rotation.
According to a report titled "Fund Flows and Liquidity: The Need for AI Rotation" released by JPMorgan Chase on July 3rd, semiconductor stocks—specifically, AI chip and memory manufacturers—have consistently and almost steadily outperformed hyperscale cloud service providers since September of last year. This performance gap appears unsustainable in the long term. The report argues that since semiconductor trading is inherently part of a broader AI market, the current divergence is raising concerns about its sustainability. JPMorgan Chase states that this gap could narrow in two ways. In a positive scenario, as hyperscale cloud providers, AI model providers, and users improve in commercialization, revenue, and profitability, their performance will begin to catch up, gaining a larger share of overall AI value-added. In a negative scenario, if the semiconductor outperformance comes at the expense of customers such as hyperscale cloud providers, AI model providers, or end users, it could dampen their willingness to invest in capital expenditures, ultimately hindering demand for semiconductor companies' products. The report notes that while JPMorgan Chase's internal view leans towards a more positive scenario, analysts' consensus forecast indicates a significant slowdown in hyperscale cloud service provider capital expenditure growth starting next year, which, if true, is closer to a negative scenario. The report states that hyperscale cloud service provider capital expenditure growth is projected to reach 100% in 2026, but may drop to 22% in 2027 and further to 7% in 2028. If this slowdown path holds true, semiconductor transactions could face significant pressure, triggering a more significant and sustained correction in AI transactions across the stock and bond markets. JPMorgan Chase also states that the price of AI computing power will be crucial for hyperscale cloud service providers to commercialize their AI capital expenditures. Higher computing power prices will enable cloud service providers to maintain or increase profit margins. Furthermore, the report states that the rate of money creation in the United States is projected to rise from $1.6 trillion in 2025 to $1.8 trillion in 2026, continuing to support US financial assets, particularly US stocks.
Analysis: The AI investment boom is cooling down, and the market is reassessing the sustainability of chip and data center spending.
According to Odaily Odaily, the AI infrastructure investment boom is cooling down, and the market is beginning to reassess the sustainability of chip and data center spending. As investors re-examine whether AI infrastructure investment can be sustained, "AI deals" covering the semiconductor, memory chip, and data center industry chain are showing signs of cooling. Recently, AI-related chip stocks such as Micron Technology (MU) and SanDisk (SNDK) have been under pressure. Samsung Electronics previously reported record second-quarter results, but revenue fell short of market expectations, causing its stock price to drop nearly 7%, dragging down the entire AI chip sector. Market concerns are growing that the current AI boom, driven by GPUs, high-bandwidth memory (HBM), and data center construction, may face repricing as cloud computing giants (Hyperscalers) may slow their investments in AI infrastructure. Meanwhile, South Korean memory chip giant SK Hynix's stock price has fallen about 25% from its all-time high ahead of its US IPO, which is also attracting some funds away from existing chip stocks. Analysts point out that after SpaceX's massive IPO boosted valuations of AI-related assets, investors are reassessing the growth logic for the next phase of the AI market. If the AI investment fervor cools further, some funds may flow back from the AI industry chain to other risky assets, including crypto assets. (CoinDesk)
The computing chip concept saw renewed activity, with CloudWalk Technology rising over 10%.
Mars Finance reported on July 8th that the computing chip concept saw renewed activity, with Cloudwalk Technology surging over 10%, Rockchip nearing its daily limit, and Fudan Microelectronics, ASR Microelectronics, Wantong Development, Espressif Systems, and VeriSilicon following suit. In terms of news, JPMorgan Chase predicts that ASIC chip shipments will accelerate significantly in the next two years. In 2026, ASIC shipments are projected to reach 6.8 million units, while GPU shipments are expected to reach 9.5 million units, with ASICs accounting for approximately 42% of global AI chip shipments. However, by 2027, ASIC shipments are projected to reach 12.5 million units, accounting for 53%, surpassing GPUs' 10.9 million units for the first time. (Science and Technology Treasure Broadcast)