Waking up, Meta's Zuckerberg did something big: announcing that Meta will officially sell its own AI cloud computing service to the outside world.
As soon as the news was announced, the chip sector in the US stock market faced a severe sell-off:
- Philadelphia Semiconductor Index: fell 6.27% in a single day
- Micron: plummeted 10.57%
- SanDisk: plummeted 10.62%
- Intel: plummeted 9.03%
- Corning: plummeted 13%
A storm of computing power initiated by Meta is pulling the entire AI hardware supply chain into an intense debate over the "ceiling of demand".
1. Meta's "Turnaround": From Burning Money to Selling Water
To understand why the market reacted so violently, we need to take a look at what Meta has been doing over the past few years.
Over the past two years, Meta has been one of the most aggressive "buyers" in the AI arms race. Spending between $125 billion and $145 billion in capital expenditures each year, fiercely procuring GPUs, network equipment, optical modules, power, and cooling facilities—not for anything else but to catch up with OpenAI and Anthropic on the large model track.
The problem is, the money was spent, but the models did not deliver results. Meta's own large models have consistently lagged behind major competitors in performance, leading to a large amount of deployed computing resources being idle or operating inefficiently.
To put it figuratively: this is a marathon of the AI era. Initially, Meta was desperately chasing on the track, but as it ran, it found itself falling further behind the competitors, so it sat by the roadside and started selling water to passersby.
At the moment it sat down, it became a "computing power seller".
The specific data illustrates its substantial assets:
- By the end of 2025, Meta's AI computing power is equivalent to about 2.5 million H100s, totaling about 2GW of power scale
- In 2026, Meta provided a capital expenditure guidance of $135 billion, corresponding to an additional 2-3GW of computing power
- Simple calculations suggest that by the end of 2026, the total computing power in hand could approach 5GW
What does this mean? It means that the AI computing capabilities owned by Meta might already exceed those of any tech company except for the hyperscale cloud providers. Furthermore, it itself cannot use it all.
Therefore, the logic of renting out idle computing power makes sense—it would be a waste to have assets sitting there unused; turning them into cash can at least make the financial statements look better.
2. Why did chip stocks fall? Two Voices in the Market
Regarding Meta's sale of computing power, there are currently two different interpretations in the market.
Interpretation 1: Bearish—The signal of "computing power surplus" has been fired
The bearish logic chain is straightforward:
Meta is one of the largest buyers of AI chips over the past two years. Now it has begun to rent out excess computing power—this indicates that it no longer needs to buy more.
If even a player of Meta's caliber begins to realize they have "overbought," what about other manufacturers? Is the arms race for large models transitioning from "full sprint" to "applying the brakes"?
Further deduction is: once a consensus of "computing power surplus" forms, the expected incremental demand for the entire AI hardware market will be significantly revised down. Key players in the supply chain such as Samsung, TSMC, Micron, and NVIDIA could face the risks of slowing order growth or even order cuts. The hardware supply chain, which has been prospering for over two years, might begin to discount.
Interpretation 2: Bullish—"Selling water" is to continue running the marathon
The bullish counterargument is equally strong:
The GPUs that Meta spent hundreds of billions on in the past two years are sunk costs that have already been spent. Now, activating and monetizing these idle assets is not abandoning competition but returning to business rationality.
If renting out computing power can indeed generate revenue, Meta will have even more confidence in subsequent purchases of GPUs, network equipment, and optical modules—because they gain backability, burning money can become more sustainable. Spending on equipment → renting out idle computing power to recoup funds → using returned funds to buy more equipment, this is actually a positive cycle, not a zero-sum game.
3. Who is the Winner? The Answer from Capital is Clear
The two views are currently in intense confrontation, and the short-term volatility of chip stocks is unlikely to calm down. But one thing is certain—the capital market is rewarding the decision of Meta to "sell water" itself with real money.
In the trading days following the announcement, Meta's stock price did not plummet along with the chip stocks. On the contrary, it surged nearly 9%. This indicates a clear investor sentiment: regardless of how chip stocks fall, this move by Meta is a plus for itself.
Why is that?
The core issue is not "how much money can renting out computing power make". Because even if Meta rents out all its excess computing power, it is hard to say how much incremental net profit it can bring in the short term. Perhaps the initial figure is only in the range of $2 billion to $3 billion, which is not significant for a company with annual revenues in the hundreds of billions.
What the market truly cares about is the change in attitude.
In the past few years, Wall Street's biggest concern about Meta has not been its GPU purchases but its "bottomless money burning". Annual capital expenditures of $125 billion to $145 billion are like a pit that can never be filled, causing investors’ doubts about return on investment (ROI) to grow increasingly larger.
Against this backdrop, Zuckerberg's willingness to monetize surplus computing power—even if the amount is small—conveys the message: the management is beginning to care about capital efficiency, and the era of "crazy money burning" may be reaching its peak.
This is the signal Wall Street has been waiting for. Therefore, even when chip stocks plummet, Meta's own stock price surged against the trend. The capital is rewarding not the computing power rental business itself but the strategic shift from "unlimited arms race" to "return to business rationality".
4. The Real "Black Swan" May Not Have Appeared Yet
Although chip stocks have fallen dramatically, to say that the AI hardware bull market has ended may still be premature.
The biggest uncertainty currently lies not only in the fact of Meta renting out computing power but also whether other giants will follow suit.
Meta is the first major tech company to publicly state "we have overbought computing power and need to rent it out". What about Microsoft and Amazon? If these cloud giants, which are also heavily procuring AI infrastructure, buckle under pressure from capital markets and announce they will "invest rationally" and reduce capital expenditure plans—that would be the moment that truly shakes the foundation of the entire AI hardware supply chain.
So far, Microsoft and Amazon have not released similar signals. However, if in the coming weeks or months, "cutting capex" becomes a collective action of tech giants, the value of AI chip hardware will be reassessed.
In other words, the current market is shifting from one question to another: the issue is not with Meta, but whether Meta could be the first domino to fall.
5. In Conclusion: Volatility is Certain, Direction is Uncertain
For investors concerned about chip stocks, the current situation can be summarized in one sentence: the uncertainty is extremely high, and volatility will likely intensify in the short term.
How significant will Meta's rental of computing power impact chip demand? Is it an isolated incident or the start of an industry trend? Will Microsoft and Amazon follow suit? These questions currently have no answers. Until answers surface, the stock prices in the chip sector are likely to continue fluctuating.
For investors looking to trade in chip stock volatility, the BIT (formerly Matrixport) platform provides access to real chip stock trading listed on NASDAQ—covering core stocks such as Micron, Intel, AMD, NVIDIA, etc. You are buying the underlying stocks registered within the DTC system, not tokenized certificates, and trades can happen anytime through ACATS transfer.
With BIT's USDT 24/7 near-instant funding channel, even if you are in the Asian time zone, you can complete fund transfers before US stock market hours without missing any important market windows.
The story of AI hardware is not over yet, but the script is being rewritten.
Disclaimer: This article is for informational reference only and does not constitute investment advice. The market interpretations and data in this article are based on public information and may have delays or discrepancies. The volatility risks of US stocks and the chip sector are high, and past performance does not guarantee future returns. The specific terms of related trading services are subject to the official explanation of the BIT platform, and users from different regions must confirm their compliance. Investment carries risks; please consult a professional advisor before making decisions and bear the related risks and consequences.
免责声明：本文章仅代表作者个人观点，不代表本平台的立场和观点。本文章仅供信息分享，不构成对任何人的任何投资建议。用户与作者之间的任何争议，与本平台无关。如网页中刊载的文章或图片涉及侵权，请提供相关的权利证明和身份证明发送邮件到support@aicoin.com，本平台相关工作人员将会进行核查。
