<div class="fs-article fs-responsive-text current-article article-body"><p>For the last few years, investors have been able to hide behind one trade. Own the Magnificent 7. Own the index. Own the future.</p><p>It worked because scale became the answer to almost every question the market cared about. Who had the strongest balance sheets? Big Tech. Who had the data? Big Tech. Who had the distribution, the engineers, the customer relationships, the platforms, and the ability to spend through any cycle? Big Tech.</p><p>Then the market structure made it even stronger. As these companies became larger, index funds had to own more of them. The companies grew even larger as index funds acquired a greater share of them. That loop was not irrational. It was not a bubble in a simple sense. It was how market capitalization weighting works when the biggest businesses also happen to be the best-performing stocks. Unfortunately, good trades eventually become bad habits.</p><section id="magnificent-7-became-substitute-analysis"><h2>The Magnificent 7 Became A Substitute For Analysis</h2><p>The Magnificent 7 label became a substitute for analysis. It allowed investors to stop asking what they owned. Someone folded growth, quality, AI, momentum, index exposure, balance sheet strength, and safety into one convenient phrase. The trade worked so well that the explanation became less important than the ownership. In my opinion, that is usually when the market starts to change. I speak a lot in my book about behavioral edge.</p><p>The Magnificent 7 is no longer one trade. They are seven very different companies, each with its own unique economic exposure. AI is making that more obvious, not less. Earnings dispersion is widening. Capital spending needs are diverging. Return on invested capital matters again. The market is beginning to separate AI as a story from AI as an economic engine. That distinction will decide the next phase of returns.</p><p>I have written about this risk before, though at the time it looked more like a concentration issue than a dispersion issue. In How Soaring Big Tech Stocks Threaten To Crash Your 401(k)<em>,</em> I warned that many investors who believed they owned a diversified index were really leaning on a small group of massive technology companies. The line was blunt: “If you invest in the S&amp;P 500, you are essentially gambling on the 7 largest tech stocks.”</p><p>That was not an argument against owning great companies. It was an argument against confusing market-cap weighting with diversification. A few months later, in Big Tech’s Overpowering Influence: Risks To Markets And Your Money, I came back to the same problem from another angle. Big Tech had become so dominant that it was not just driving returns. It was shaping how investors thought about risk. The index still looked diversified on a fact sheet, but the return stream was becoming narrower underneath the surface.</p></section><section id="magnificent-7-no-longer-moving"><h2>The Magnificent 7 Is No Longer Moving As One Trade</h2><p>That matters now because the trade is no longer moving like one clean block. The first phase rewarded exposure. You did not have to be especially subtle. If AI, cloud, chips, digital advertising, software, platforms, and automation were the future, the answer was to own the companies most clearly attached to that future. That worked. The next phase will be harder.</p><p>Now investors have to ask a more uncomfortable question. Who benefits from AI spending, and who is being forced to spend just to defend an existing profit pool?</p><p>There is a major difference between AI capital spending that deepens a moat and AI capital spending that becomes a tax. A company that can invest in AI and convert that spending into better products, higher retention, stronger pricing power, and deeper customer dependency is in a powerful position. That capital is offensive. It expands the advantage.</p><p>But a company that must spend because not spending is too dangerous is in a different position. The company may still be excellent. It may still produce enormous cash flow. It may still dominate its category. But if the next decade requires materially higher infrastructure spending, more depreciation, heavier compute needs, and lower incremental returns, the old valuation framework may not survive untouched.</p><p>The market cannot value a business as capital-light forever if the business starts behaving like a capital-heavy one. That is the real issue under the AI boom.</p></section><section id="ai-separating-magnificent-7-winners"><h2>AI Is Separating The Magnificent 7 Winners From The Spenders</h2><p>This point is where the Magnificent 7 begins to separate. Microsoft is not Apple. Apple is not Nvidia. Nvidia is not Tesla. Tesla is not Meta. Meta is not Alphabet. Alphabet is not Amazon. They may sit together in the same market basket, but their economics are not the same.</p><p>I wrote in 2023 that Microsoft Stock Is Set To Skyrocket Into The AI Future because Microsoft had more than a remarkable AI story. It had distribution, cloud scale, enterprise trust, customer dependency, and the ability to place AI directly inside workflows companies already used. That is not hype. That is structure. AI attached to an existing enterprise toll road is different from AI attached to a promise.</p><p>That is why the market is getting more selective.</p><p>Nvidia is still the clearest infrastructure winner because it sells what others need to build. Microsoft has the advantage of embedding AI into software and cloud relationships that already exist. Amazon has the cloud scale but must prove the returns on the spending. Alphabet has extraordinary AI capability, but AI also raises questions about the economics of search. Meta has shown that AI can improve advertising efficiency and engagement, but investors remember what happens when spending gets too far ahead of visible returns. Apple has the ecosystem but still needs to prove AI can drive a new consumer cycle rather than simply protect the existing one. Tesla has autonomy and robotics optionality, but optionality is not the same as current earnings power. Same label. Different businesses. That is why the trade is breaking apart.</p><p>Not because these companies are suddenly weak. They are not. But the market is moving from admiration to underwriting. It is no longer enough to say a company is exposed to AI. Investors now have to ask whether AI improves the business model or absorbs more of the cash flow.</p></section><section id="structural-alpha-where-next-value"><h2>Structural Alpha Is Where The Next Value Is Hiding</h2><p>This area is also where structural alpha becomes more important.</p><p>Structural alpha is not about finding a clever opinion on the same seven stocks that everyone already owns. It is about finding the places where market structure, capital flows, incentives, and forced behavior create mispricing. Most investors spend their time asking what people will think next. I would rather ask what people will be forced to do next. That is usually where the better setup sits. A company has to separate a business. A board has to respond to pressure. An index fund has to sell because a company no longer fits the mandate. A management team has to simplify. A balance sheet has to be repaired. A data center has to secure power. A platform has to buy compute. A utility has to upgrade the grid. A supplier has to expand capacity because demand has moved from optional to necessary. These are not opinions. They are actions. And when action becomes unavoidable, price often follows. The market does not always reprice because investors suddenly discover something new. Often it reprices because the structure changes and investors are forced to look at the same assets differently.</p></section><section id="ge-western-digital-show-pattern"><h2>GE And Western Digital Show The Pattern</h2><p>That was the lesson from GE.</p><p>GE was not unknown. It may have been one of the most analyzed companies in the world. Investors knew the mistakes, the pension issues, the bad acquisitions, the management failures, and the destroyed trust. What they did not fully price was the structural change. Once GE broke itself apart, we could finally judge the pieces on their own terms. Aviation was no longer buried inside a damaged conglomerate. Healthcare had its own investor base. GE Vernova became a focused way to own electrification, gas turbine demand, grid investment and power scarcity just as those issues became central to the AI infrastructure buildout.</p><p>The alpha did not come from discovering GE. It came from understanding that the old structure was hiding the value of the pieces.</p><p>Western Digital was a similar case in a different wrapper. The market saw a cyclical storage company with weak sentiment and a messy mix of businesses. But the SanDisk separation changed the frame. It changed the shareholder base, the narrative, and the valuation lens. AI and data growth made storage more strategically relevant at the same time the corporate structure was being simplified.</p><p>Again, the point was not simply that storage was an AI beneficiary. The point was that the structure changed as the demand backdrop improved. That is when markets often misprice things.</p><p>I wrote about this broader idea in Companies Are Breaking Up. Investors Can Gain. These 3 Tech Giants Are Next. Large companies often hide value in the wrong corporate wrapper. The market may understand the assets but still misprice the structure. When the wrapper changes, the valuation can change quickly.</p><p>The Magnificent 7 breakup is not the same thing as a corporate breakup, but the logic is related. The market has been valuing a group. Now it has to value individual businesses. That creates risk for passive investors, but it also creates opportunity for active ones.</p></section><section id="better-ai-trade-may-be"><h2>The Better AI Trade May Be Beneath The Magnificent 7</h2><p>If the mega-cap platforms are spending hundreds of billions of dollars on AI infrastructure, the obvious trade is to own the platforms. The more interesting trade is to follow where the spending has to go. Power, grid equipment, cooling, memory, storage, semiconductor equipment, fiber, networking, data center real estate, engineering, select industrials, and certain utilities are no longer just background industries. They are becoming part of the AI supply chain.</p><p>Some of these businesses are still valued like old economy assets. That is the opportunity.</p><p>The market loves the clean story. It loves the company with the obvious AI headline, the liquid stock, the sell-side coverage, and the easy narrative. But the better return is often found in the bottleneck. The supplier. The forced beneficiary. The asset that becomes more valuable because someone else has no choice but to spend.</p><p>That is what investors should be looking for now.</p><p>Not a blind rotation out of technology. That is too crude. Not a knee-jerk sale of the winners because they have gone up. That is not analysis either. The better question is where the AI capital cycle is changing the economics of businesses the market still thinks it understands. That is how structural alpha usually appears. Quietly first. Then suddenly obvious.</p><p>A utility with exposure to data center demand may still be valued like a slow-growth regulated asset until power scarcity changes the conversation. An equipment supplier may look cyclical until orders become tied to years of infrastructure investment. A spin-off may look too small or too messy for the parent company’s shareholder base until forced selling clears and the right owners show up. A breakup may look like financial engineering until the separated assets attract investors who would never have owned the old conglomerate.</p><p>These are not glamorous setups. That is partly why they work. The best opportunities rarely arrive with perfect narratives. They usually arrive with some discomfort attached. A messy shareholder base. A forced seller. A complicated capital structure. An unloved asset. A board under pressure. A business that screens poorly until the structure changes is a problem.</p><p>That is where I would rather spend time than chasing the most crowded version of the AI trade.</p></section><section id="what-magnificent-7-breakup-means"><h2>What The Magnificent 7 Breakup Means For Investors</h2><p>The S&amp;P 500 may still look diversified, but investors should be honest about what they own. When a handful of stocks drive a large share of returns, the index becomes more exposed to the same assumptions. That was manageable when the Magnificent 7 moved together. It becomes more dangerous when dispersion widens.</p><p>Market-cap weighting is not risk control. It is a rules-based way of owning more of what has already gone up. That can work beautifully for long periods. But when leadership begins to separate, investors need to stop treating the index as if it automatically solves the problem.</p><p>Dispersion is not bad for serious investors. It is healthy. It means work matters again. It means the market is no longer rewarding every version of the same story equally. It means lazy baskets get punished and better underwriting gets rewarded.</p><p>The Mag 7 trade worked because concentration was rewarded. That is not a permanent law of markets. It is a condition, and conditions can change. That is a better environment for active investors, but not an easier one. The easy money was made by owning the label. The next money will be made by understanding the structure beneath it.</p><p>Which companies are earning high returns on AI capital? Which ones are spending defensively? Which suppliers are becoming mission-critical? Which old economy assets are being pulled into a new demand cycle? Which spin-offs, breakups, and special situations are being ignored because they do not fit neatly into the AI narrative? Those are the questions that matter now.</p><p>The Magnificent 7 may still dominate the index. It no longer dominates the opportunity set. The next phase of the market will not emerge if we move away from technology. You will find it by looking beneath it. That is where capital is flowing, bottlenecks are forming, and the next structural alpha is likely to appear.</p></section></div>

1.25 倍做多特斯拉 ETF - GraniteShares

特斯拉

<p>“七大巨頭” 交易使投資者能夠依賴少數主導科技股，但這一趨勢正開始分化。隨着人工智能影響市場動態，這些公司的獨特經濟暴露逐漸顯現。投資者現在必須辨別哪些公司將從人工智能投資中受益，以及哪些公司可能面臨增加的資本支出負擔。這一轉變表明，需要更細緻的分析，因為之前對這些股票的普遍投資方法可能不再適用</p>

The Magnificent 7 Trade Is Starting To Break Apart

- 投資者此前依賴大型科技股，市場高度集中導致風險隱患增加。  
- 目前 “壯觀七巨頭” 逐漸分化，獨立公司各自面臨不同挑戰與機遇。  
- 下階段的投資需關注更深層次的行業變化與新興機會，而非單一巨頭表現。

壯麗七巨頭的交易開始出現分裂