The DRAM ETF pulled back about 25% from its June 22nd high, the semiconductor ETF (SMH) lost 12% in two weeks, and the Philadelphia Semiconductor Index has faced consecutive corrections. Micron delivered earnings that could be described as “explosive” — revenue of $41.4 billion and forward guidance pointing towards $50 billion — yet its stock price reversed downwards on a “buy the rumor, sell the news” logic.
And Meta, arguably the most aggressive 人工智能 infrastructure buyer over the past two years, officially announced last week that it will rent out its spare computing power to external customers, which served as the direct catalyst for the chip sector’s pullback.
Even the most fervent “GPU hoarders” are starting to sell off their idle inventory.
This signal quickly triggered a chain reaction in the market. Following this, Samsung’s earnings report further solidified investor concerns: the surge in profits for the memory chip industry essentially means the entire sector is riding the same super-cycle wave, rather than any single company monopolizing an irreplicable moat. Many investors are already starting to worry — is the AI narrative coming to an end?
1. This Might Be a “Rotation,” Not an “End”
Amidst the widespread bearish sentiment on chips, one of Wall Street’s most influential strategists offered a starkly different assessment.
Morgan Stanley’s Chief US Equity Strategist, Michael Wilson, stated directly in his latest weekly report: underweight semiconductors, rotate into hyperscale cloud computing vendors.
The weight of this statement lies in the fact that it isn’t declaring “AI is over,” but rather saying “the direction of profit distribution has shifted.”
Wilson’s analytical logic is this: over the past two years, the most profitable part of the AI value chain has been the “pick-and-shovel sellers” — Nvidia, Micron, SK Hynix. They provide the underlying computing and storage infrastructure for the AI revolution, enjoying pricing power and ultra-high margins driven by supply-demand imbalances. However, as cloud providers’ capital expenditure growth approaches an inflection point, the growth rate of “shovel” demand is transitioning from a boom phase to a stabilization phase.
Going forward, the profit center of the value chain is shifting from the hands of the “shovel sellers” to the “miners using the shovels.”
Cloud providers — such as Microsoft, Google, Amazon, and Alibaba — are the ultimate integrators and commercializers of AI capabilities. They use chips to build computing power, package that power into cloud services, AI assistants, and enterprise solutions, and sell them to millions of users and businesses globally. As the price hike cycle for upstream chips begins to slow, the cost pressure on cloud providers actually eases, while their AI service revenue continues to climb.
This marks a peak in the *rate* of adjustment, not the peak of the entire capital expenditure cycle. In other words, cloud providers won’t stop investing in AI, but the fastest phase of investment growth is likely over. Consequently, the valuation of chip stocks needs to transition from an “explosive growth” pricing model to a “mature growth” pricing model — and this transition process is often accompanied by significant capital migration and stock price volatility.
2. Memory Chips Fall First: A Potential Start to Capital Rotation
Let’s look at how the data validates this logic.
Since Meta announced its plan to rent out computing power, memory chips have become the epicenter of the correction storm — because memory demand is most directly dependent on the capital expenditure intentions of cloud providers. When the biggest buyers start signaling “we’ve bought enough,” cracks appear in the growth narrative for memory suppliers.
Micron’s case is the most typical. The company reported Q3 revenue of $41.4 billion and management’s forward guidance was as high as $50 billion — a “killer hand” in any normal environment. Yet the market’s response was a stock price decline. When “beating expectations” becomes the norm, what truly determines the stock price is no longer the numbers themselves, but “can they continue to exceed expectations?” This “sell the news” price action is itself a characteristic of capital pricing a cycle top.
But if the AI cycle were truly ending, what should we be seeing? We should see a comprehensive collapse of all AI-related assets — cloud providers, AI application companies, Chinese tech stocks, none spared.
But what the market presents is a different picture: capital is flowing *out* of chips, but *into* another AI narrative.
3. Alibaba Surges 11%: A Signal Flare of the New Rotation
One of the most compelling pieces of evidence occurred during US trading hours yesterday.
Alibaba’s US-listed shares surged 11% in a single day. Prior to this, the chip sector was experiencing a brutal sell-off. If the AI narrative were truly broken, Alibaba, as a downstream player in the AI value chain, should logically have fallen in tandem. But the opposite happened — capital retreated from chips and flowed into Chinese cloud computing and AI platform companies represented by Alibaba.
Two layers of logic resonate behind this.
The first logic is the rotation itself. As Wilson stated, the profit center of the value chain is migrating from hardware infrastructure to software platforms and cloud services. Alibaba, as China’s largest cloud provider and one of the leading AI large model developers, is positioned to benefit from this shift in profit distribution.
The second layer of logic involves geopolitics. There are market rumors (unconfirmed by official channels as of press time) suggesting the Chinese government plans to impose stricter restrictions on the export and overseas access to cutting-edge AI models. Such a policy signal indicates that the AI competition between the US and China is intensifying further — with each nation trying to build its own sovereign and controllable AI technology stack. Against this backdrop, local platform companies possessing a complete AI ecosystem (cloud computing, large models, application scenarios) are having their strategic value reassessed by the capital market.
It’s worth noting that US-based cloud providers — Microsoft, Google, Amazon — have yet to show a similarly drastic positive reaction in their recent stock performance as Alibaba. This could be because their valuations were already at elevated levels, or because the market is still digesting the suppressive effect of “peaking capital expenditure growth” on short-term profit margins. However, based on the rotation logic, if capital is indeed shifting from chips to cloud platforms, a valuation recovery for these giants is likely just a matter of time.
4. Final Thoughts
The correction in chip stocks is painful, but the AI story is far from over.
Only the next chapter’s protagonists may no longer be those selling GPUs and HBM, but those using GPUs and HBM to build the next generation of the internet. Capital is flowing from the hands of the “shovel sellers” to the “miners using the shovels” — this is not a funeral for AI, but an internal profit redistribution within the AI value chain.
On the BIT (formerly Matrixport) platform, you can trade not only chip stocks like Micron and Nvidia, but also directly invest in real US stocks of cloud platform companies like Alibaba, Microsoft, Google, and Amazon. Covering over 10,000 US stocks and ETFs, it allows you to quickly adjust your portfolio whenever a rotation occurs.
[Disclaimer] This article is market observation and commentary content, representing the author’s analysis of public information and personal views. It does not constitute investment advice or a recommendation or solicitation for the purchase or sale of any securities. The views of third-party institutions (including Morgan Stanley and its analyst Michael Wilson) cited in this article represent only the stance of the individuals or their organizations and do not represent the views or judgments of BIT. 市场 trends and individual stock performance are influenced by multiple factors. The causal analysis in this article is only one possible interpretation and is not guaranteed for accuracy, nor does it guarantee that future markets will follow this logic. Investing involves the risk of loss of principal. Please make prudent decisions.
本文来源于互联网： Chip stocks have cooled off, but the AI cycle may not be dead yet
1. Introduction 2. The Tornado Cash Case: A Textbook Example of Regulatory Overreach The Tornado Cash case is the most important on-chain enforcement precedent of the past four years. The sanctions imposed in August 2022 sent shockwaves through the industry: GitHub closed the code repository, Circle froze USDC addresses that had interacted with Tornado Cash, and Uniswap blocked related trading pairs on its front end—yet the underlying contracts remained entirely unaffected. The power of a single executive order was completely deconstructed by a line of immutable code. OFAC’s enforcement assumption was based on a fundamental miscalculation: believing that “freezing the front end” is equivalent to “freezing the protocol.” The result proved these are two different things—sanctions lists are compliance inventories, not physical bans. Front-end service providers will comply, but blockchain…
简体中文 
English 
繁體中文