<p>邁克爾·伯裏本週在對人工智能（AI）交易進行做空時設定了入場價格。在他 6 月 30 日的 Substack 帖子中，這位因《大空頭》而聞名的對沖基金經理披露了對 <strong>英偉達</strong>（NVDA 1.39%）、<strong>特斯拉</strong>（TSLA 7.35%）、<strong>應用材料</strong>（AMAT 7.51%）、<strong>卡特彼勒</strong>（CAT 2.66%）和 <strong>iShares 半導體 ETF</strong>（SOXX 5.57%）的空頭頭寸。兩天後，他又對 <strong>美光科技</strong>（MU 5.68%）進行了做空，報道稱其入場價接近每股 1,052 美元。</p>
<p>伯裏是那位在 2008 年崩盤前對房地產市場做空的投資者。當他同時針對市場上三隻最廣泛持有的股票時，問題幾乎自然而然地出現：你應該擔心嗎？</p>
<p>答案首先是認真對待他的論點。</p>
<p><img src="https://imageproxy.pbkrs.com/https://g.foolcdn.com/image//query-b3A9cmVzaXplJnVybD1odHRwczovL2cuZm9vbGNkbi5jb20vZWRpdG9yaWFsL2ltYWdlcy84Nzc0NzQvc3RvY2stcHJpY2UtZmFsbGluZy5qcGcmdz0zODQw?x-oss-process=image/auto-orient,1/interlace,1/resize,w_1440,h_1440/quality,q_95/format,jpg" original-src="https://imageproxy.pbkrs.com/https://g.foolcdn.com/image//query-b3A9cmVzaXplJnVybD1odHRwczovL2cuZm9vbGNkbi5jb20vZWRpdG9yaWFsL2ltYWdlcy84Nzc0NzQvc3RvY2stcHJpY2UtZmFsbGluZy5qcGcmdz0zODQw"/></p>
<p>圖片來源：Getty Images。</p>
<h2>伯裏的論點</h2>
<p>伯裏的論點並不是這些企業正在失敗，而是它們的股價已經達到歷史極端。他據稱指出，美光現在的交易價格遠高於其 200 日移動平均線，這是自 1984 年以來的最高點，這一時期包括了互聯網泡沫的巔峯。</p>
<p>歷史在他的論點中起到了很大作用。美光已經上市四十年，根據伯裏的統計，其股票在此期間經歷了 34 次超過 30% 的回調。在他看來，AI 熱潮並沒有取消內存週期，而是讓投資者忘記了這個週期的存在。</p>
<p>其餘的做空組合在 AI 供應鏈中延續了相同的邏輯：應用材料銷售用於製造先進芯片的設備，卡特彼勒提供數據中心建設所需的機械和電力系統，而半導體 ETF 則將整個行業打包成一個單一的股票代碼。</p>
<p>當然，一個著名的名字和一張令人擔憂的圖表並不能使論點正確。投資者已經對 AI 交易稱之為泡沫超過兩年，而這些股票仍在上漲。更有用的問題是這三隻股票各自有多依賴伯裏所攻擊的樂觀情緒。</p>
<h2>三隻股票，三種不同的做空</h2>
<p>美光是他論點中最有趣的，因為從今天的數字來看，這隻股票並不顯得昂貴。在其 2026 財年第三季度（截至 2026 年 5 月 28 日），這家內存製造商的收入同比增長超過四倍，達到 415 億美元，較上一季度增長約 74%。管理層預計 2026 財年第四季度的收入大約為 500 億美元。</p>
<p>展開</p>
<p><img src="https://imageproxy.pbkrs.com/https://g.foolcdn.com/image//query-b3A9cmVzaXplJnVybD1odHRwczovL2cuZm9vbGNkbi5jb20vYXJ0L2NvbXBhbnlsb2dvcy9tYXJrL01VLnBuZyZ3PTEyOA?x-oss-process=image/auto-orient,1/interlace,1/resize,w_1440,h_1440/quality,q_95/format,jpg" alt="美光科技股票報價" original-src="https://imageproxy.pbkrs.com/https://g.foolcdn.com/image//query-b3A9cmVzaXplJnVybD1odHRwczovL2cuZm9vbGNkbi5jb20vYXJ0L2NvbXBhbnlsb2dvcy9tYXJrL01VLnBuZyZ3PTEyOA"/></p>
<h2>NASDAQ: MU</h2>
<p>美光科技</p>
<p>今日變動</p>
<p>(-5.68%) $-58.69</p>
<p>當前價格</p>
<p>$973.59</p>
<h3>關鍵數據點</h3>
<p>市值</p>
<p>$1.1 萬億</p>
<p>日內區間</p>
<p>$950.30 - $1064.56</p>
<p>52 周區間</p>
<p>$103.38 - $1255.00</p>
<p>成交量</p>
<p>1.7M</p>
<p>平均成交量</p>
<p>51.3M</p>
<p>毛利率</p>
<p>72.60%</p>
<p>股息收益率</p>
<p>0.05%</p>
<p>截至目前，美光的股價約為每股 976 美元，市盈率大約為 22 倍。這看起來很適中。但內存高峰通常都是如此——當收益激增時，市盈率會壓縮，爭論從來不是關於市盈率，而是關於如此高的收益能否持續。伯裏押注它們不能。美光的反駁是，其新的多年客户協議使需求比以往週期更持久和可預測。</p>
<p>英偉達 可能是名單上最奇怪的名字，因為其估值可以説是最容易辯護的。這家芯片製造商在 2026 財年第一季度（截至 2026 年 4 月 26 日）收入同比增長 85%，達到 816 億美元，數據中心收入增長 92%。然而，該股票的市盈率約為 30 倍。如果 AI 交易是泡沫，那麼其最大成員的定價則顯得温和得多。</p>
<p>而 特斯拉 則是相反的情況。這家電動車製造商在第二季度交付了 480,126 輛汽車，同比增長約 25%——但在報告當天，股價仍下跌了約 7.5%。但市場的關注點已經從銷量轉向每輛車的盈利能力。特斯拉第一季度的營業利潤率僅為 4.2%，在過去 12 個月的每股收益為 1.10 美元的情況下，股價的市盈率超過 350 倍。伯裏在這裏不需要週期論證。他只需要特斯拉的機器人出租車和軟件利潤的到來晚於估值的假設。</p>
<p>那麼，你應該擔心嗎？</p>
<p>並不是關於披露本身。做空者宣佈其頭寸並不會改變這些企業的任何事情。而且伯裏在重大判斷上也曾早於市場。即使他著名的房地產押注也花了幾年才見效。</p>
<p>但他的壓力測試對某些股票的影響比其他股票更大。英偉達的數字最能回應他，因為 85% 的增長和 30 倍的市盈率為其他兩者提供了沒有的緩衝。</p>
<p>其他股票則可能更難辯護。美光要求投資者相信內存定價能夠維持在行業歷史上難以維持的水平。而特斯拉則要求投資者支付超過 350 倍的市盈率，而該公司的盈利版本仍然大部分在前面。</p>
<p>總體而言，我不會僅僅因為邁克爾·伯裏對某隻股票做空就賣出該股票。但他的做空是一個誠實審視持倉規模的有用提示。如果內存下滑或機器人出租車時間表的延遲會對投資組合造成嚴重損害，那麼在論點仍未解決時考慮並可能解決這一風險是值得的，而不是等到事情結束後（可能為時已晚）。</p>

特斯拉

美光科技

英偉達

<p>對沖基金經理邁克爾·伯裏披露了對英偉達、特斯拉、美光科技、應用材料、卡特彼勒以及 iShares 半導體 ETF 的空頭頭寸，押注於 AI 交易的下跌。他認為這些股票由於歷史極端的拉伸和週期性風險而被高估，尤其是美光科技和特斯拉。儘管英偉達的強勁增長提供了估值緩衝，但文章建議投資者應考慮伯裏對內存週期和特斯拉高倍數的警告，儘管不一定僅僅基於他的行動而出售</p>

<p>Michael Burry spent this week putting entry prices on a bet against the artificial intelligence (AI) trade. In a June 30 post on his Substack, the hedge fund manager made famous by &#34;The Big Short&#34; disclosed short positions in <strong>Nvidia</strong>( NVDA 1.39%), <strong>Tesla</strong>( TSLA 7.35%), <strong>Applied Materials</strong>( AMAT 7.51%), <strong>Caterpillar</strong>( CAT 2.66%), and the <strong>iShares Semiconductor ETF</strong>( SOXX 5.57%). Two days later, he added a short against <strong>Micron Technology</strong>( MU 5.68%), reportedly entered near $1,052 per share.</p>
<div class="lb-trans"><p>邁克爾·伯裏本週在對人工智能（AI）交易進行做空時設定了入場價格。在他 6 月 30 日的 Substack 帖子中，這位因《大空頭》而聞名的對沖基金經理披露了對 <strong>英偉達</strong>（NVDA 1.39%）、<strong>特斯拉</strong>（TSLA 7.35%）、<strong>應用材料</strong>（AMAT 7.51%）、<strong>卡特彼勒</strong>（CAT 2.66%）和 <strong>iShares 半導體 ETF</strong>（SOXX 5.57%）的空頭頭寸。兩天後，他又對 <strong>美光科技</strong>（MU 5.68%）進行了做空，報道稱其入場價接近每股 1,052 美元。</p>
</div><p>Burry is the investor who bet against the housing market ahead of the 2008 crash. When he targets three of the market's most widely held stocks at once, the question practically asks itself: Should you be worried?</p>
<div class="lb-trans"><p>伯裏是那位在 2008 年崩盤前對房地產市場做空的投資者。當他同時針對市場上三隻最廣泛持有的股票時，問題幾乎自然而然地出現：你應該擔心嗎？</p>
</div><p>The answer starts with taking his argument seriously.</p>
<div class="lb-trans"><p>答案首先是認真對待他的論點。</p>
</div><p><img src="https://imageproxy.pbkrs.com/https://g.foolcdn.com/image//query-b3A9cmVzaXplJnVybD1odHRwczovL2cuZm9vbGNkbi5jb20vZWRpdG9yaWFsL2ltYWdlcy84Nzc0NzQvc3RvY2stcHJpY2UtZmFsbGluZy5qcGcmdz0zODQw" alt="A chart showing a stock price declining." original-src="https://g.foolcdn.com/image/?url=https%3A%2F%2Fg.foolcdn.com%2Feditorial%2Fimages%2F877474%2Fstock-price-falling.jpg&amp;w=3840&amp;op=resize"/></p>
<p>Image source: Getty Images.</p>
<div class="lb-trans"><p>圖片來源：Getty Images。</p>
</div><h2>The case Burry is making</h2>
<div class="lb-trans"><h2>伯裏的論點</h2>
</div><p>Burry's argument isn't that these businesses are failing. It's that their stock prices have stretched to historical extremes. Micron, he reportedly noted, now trades further above its 200-day moving average than at any point since 1984 -- a stretch that includes the dot-com peak.</p>
<div class="lb-trans"><p>伯裏的論點並不是這些企業正在失敗，而是它們的股價已經達到歷史極端。他據稱指出，美光現在的交易價格遠高於其 200 日移動平均線，這是自 1984 年以來的最高點，這一時期包括了互聯網泡沫的巔峯。</p>
</div><p>History does much of the work in his thesis. Micron has been public for four decades, and by Burry's reported count, its stock has suffered 34 drawdowns of more than 30% along the way. In his view, the AI boom hasn't repealed the memory cycle. It has simply made investors forget the cycle exists.</p>
<div class="lb-trans"><p>歷史在他的論點中起到了很大作用。美光已經上市四十年，根據伯裏的統計，其股票在此期間經歷了 34 次超過 30% 的回調。在他看來，AI 熱潮並沒有取消內存週期，而是讓投資者忘記了這個週期的存在。</p>
</div><p>The rest of the short book extends the same logic across the AI supply chain: Applied Materials sells the equipment used to make advanced chips, Caterpillar supplies machinery and power systems for the data center build-out, and the semiconductor ETF wraps the whole group into a single ticker.</p>
<div class="lb-trans"><p>其餘的做空組合在 AI 供應鏈中延續了相同的邏輯：應用材料銷售用於製造先進芯片的設備，卡特彼勒提供數據中心建設所需的機械和電力系統，而半導體 ETF 則將整個行業打包成一個單一的股票代碼。</p>
</div><p>Of course, a famous name and an alarming chart don't make a thesis correct. Investors have been calling the AI trade a bubble for more than two years while these stocks kept climbing. The more useful question is how much each of the three stocks depends on the optimism Burry is attacking.</p>
<div class="lb-trans"><p>當然，一個著名的名字和一張令人擔憂的圖表並不能使論點正確。投資者已經對 AI 交易稱之為泡沫超過兩年，而這些股票仍在上漲。更有用的問題是這三隻股票各自有多依賴伯裏所攻擊的樂觀情緒。</p>
</div><h2>Three stocks, three different bets</h2>
<div class="lb-trans"><h2>三隻股票，三種不同的做空</h2>
</div><p>Micron is where his case is most interesting, because the stock doesn't look expensive on today's numbers. In its fiscal third quarter (the period ended May 28, 2026), the memory maker's revenue more than quadrupled year over year to $41.5 billion, and it rose about 74% from the prior quarter. Management guided for roughly $50 billion in revenue in fiscal Q4.</p>
<div class="lb-trans"><p>美光是他論點中最有趣的，因為從今天的數字來看，這隻股票並不顯得昂貴。在其 2026 財年第三季度（截至 2026 年 5 月 28 日），這家內存製造商的收入同比增長超過四倍，達到 415 億美元，較上一季度增長約 74%。管理層預計 2026 財年第四季度的收入大約為 500 億美元。</p>
</div><p>Expand</p>
<div class="lb-trans"><p>展開</p>
</div><p><img src="https://imageproxy.pbkrs.com/https://g.foolcdn.com/image//query-b3A9cmVzaXplJnVybD1odHRwczovL2cuZm9vbGNkbi5jb20vYXJ0L2NvbXBhbnlsb2dvcy9tYXJrL01VLnBuZyZ3PTEyOA" alt="Micron Technology Stock Quote" original-src="https://g.foolcdn.com/image/?url=https%3A%2F%2Fg.foolcdn.com%2Fart%2Fcompanylogos%2Fmark%2FMU.png&amp;w=128&amp;op=resize"/></p>
<h2>NASDAQ: MU</h2>
<div class="lb-trans"><h2>NASDAQ: MU</h2>
</div><p>Micron Technology</p>
<div class="lb-trans"><p>美光科技</p>
</div><p>Today's Change</p>
<div class="lb-trans"><p>今日變動</p>
</div><p>(-5.68%) $-58.69</p>
<div class="lb-trans"><p>(-5.68%) $-58.69</p>
</div><p>Current Price</p>
<div class="lb-trans"><p>當前價格</p>
</div><p>$973.59</p>
<div class="lb-trans"><p>$973.59</p>
</div><h3>Key Data Points</h3>
<div class="lb-trans"><h3>關鍵數據點</h3>
</div><p>Market Cap</p>
<div class="lb-trans"><p>市值</p>
</div><p>$1.1T</p>
<div class="lb-trans"><p>$1.1 萬億</p>
</div><p>Day's Range</p>
<div class="lb-trans"><p>日內區間</p>
</div><p>$950.30 - $1064.56</p>
<div class="lb-trans"><p>$950.30 - $1064.56</p>
</div><p>52wk Range</p>
<div class="lb-trans"><p>52 周區間</p>
</div><p>$103.38 - $1255.00</p>
<div class="lb-trans"><p>$103.38 - $1255.00</p>
</div><p>Volume</p>
<div class="lb-trans"><p>成交量</p>
</div><p>1.7M</p>
<div class="lb-trans"><p>1.7M</p>
</div><p>Avg Vol</p>
<div class="lb-trans"><p>平均成交量</p>
</div><p>51.3M</p>
<div class="lb-trans"><p>51.3M</p>
</div><p>Gross Margin</p>
<div class="lb-trans"><p>毛利率</p>
</div><p>72.60%</p>
<div class="lb-trans"><p>72.60%</p>
</div><p>Dividend Yield</p>
<div class="lb-trans"><p>股息收益率</p>
</div><p>0.05%</p>
<div class="lb-trans"><p>0.05%</p>
</div><p>At about $976 per share as of this writing, Micron trades at roughly 22 times earnings. That looks modest. But memory peaks usually do -- the multiple compresses as earnings spike, and the argument is never about the multiple. It's about whether earnings this high can last. Burry is betting they can't. Micron's counterargument is that its new multi-year customer agreements make demand more durable and predictable than in past cycles.</p>
<div class="lb-trans"><p>截至目前，美光的股價約為每股 976 美元，市盈率大約為 22 倍。這看起來很適中。但內存高峰通常都是如此——當收益激增時，市盈率會壓縮，爭論從來不是關於市盈率，而是關於如此高的收益能否持續。伯裏押注它們不能。美光的反駁是，其新的多年客户協議使需求比以往週期更持久和可預測。</p>
</div><p>Nvidia may be the strangest name on the list, because its valuation is arguably the easiest to defend. Revenue in the chipmaker's fiscal first quarter (ended April 26, 2026) rose 85% year over year to $81.6 billion, with data center revenue climbing 92%. Yet the stock trades at about 30 times earnings. If the AI trade is a bubble, its largest member is priced like something much tamer.</p>
<div class="lb-trans"><p>英偉達 可能是名單上最奇怪的名字，因為其估值可以説是最容易辯護的。這家芯片製造商在 2026 財年第一季度（截至 2026 年 4 月 26 日）收入同比增長 85%，達到 816 億美元，數據中心收入增長 92%。然而，該股票的市盈率約為 30 倍。如果 AI 交易是泡沫，那麼其最大成員的定價則顯得温和得多。</p>
</div><p>Tesla, meanwhile, is the opposite case. The electric-car maker delivered 480,126 vehicles in the second quarter, up about 25% year over year -- and the stock still fell about 7.5% the day of the report. But the market's focus has moved past volume to what each vehicle earns. Tesla's first-quarter operating margin was just 4.2%, and with earnings per share of $1.10 over the past 12 months, the stock trades at more than 350 times earnings. Burry doesn't need a cycle argument here. He just needs Tesla's robotaxi and software profits to arrive later than the valuation assumes.</p>
<div class="lb-trans"><p>而 特斯拉 則是相反的情況。這家電動車製造商在第二季度交付了 480,126 輛汽車，同比增長約 25%——但在報告當天，股價仍下跌了約 7.5%。但市場的關注點已經從銷量轉向每輛車的盈利能力。特斯拉第一季度的營業利潤率僅為 4.2%，在過去 12 個月的每股收益為 1.10 美元的情況下，股價的市盈率超過 350 倍。伯裏在這裏不需要週期論證。他只需要特斯拉的機器人出租車和軟件利潤的到來晚於估值的假設。</p>
</div><p>So, should you be worried?</p>
<div class="lb-trans"><p>那麼，你應該擔心嗎？</p>
</div><p>Not about the disclosure itself. A short-seller announcing his positions changes nothing about these businesses. And Burry has been early on big calls before. Even his famous housing bet took years to pay off.</p>
<div class="lb-trans"><p>並不是關於披露本身。做空者宣佈其頭寸並不會改變這些企業的任何事情。而且伯裏在重大判斷上也曾早於市場。即使他著名的房地產押注也花了幾年才見效。</p>
</div><p>But his stress test lands harder on some of these stocks than others. Nvidia's numbers do the most to answer him, since 85% growth at 30 times earnings leaves a cushion the other two don't have.</p>
<div class="lb-trans"><p>但他的壓力測試對某些股票的影響比其他股票更大。英偉達的數字最能回應他，因為 85% 的增長和 30 倍的市盈率為其他兩者提供了沒有的緩衝。</p>
</div><p>The other stocks are arguably harder to defend. Micron asks investors to believe memory pricing can hold at levels the industry has historically struggled to sustain through a full cycle. And Tesla asks them to pay more than 350 times earnings, while the profitable version of the company remains mostly ahead of it.</p>
<div class="lb-trans"><p>其他股票則可能更難辯護。美光要求投資者相信內存定價能夠維持在行業歷史上難以維持的水平。而特斯拉則要求投資者支付超過 350 倍的市盈率，而該公司的盈利版本仍然大部分在前面。</p>
</div><p>Overall, I wouldn't sell a stock just because Michael Burry is betting against it. But his shorts are a useful prompt for an honest look at position sizes. If a memory downturn or a slipped robotaxi timeline would do serious damage to a portfolio, that risk is worth considering and potentially even addressing while the argument is still unsettled, not after it's over (and potentially too late).</p>
<div class="lb-trans"><p>總體而言，我不會僅僅因為邁克爾·伯裏對某隻股票做空就賣出該股票。但他的做空是一個誠實審視持倉規模的有用提示。如果內存下滑或機器人出租車時間表的延遲會對投資組合造成嚴重損害，那麼在論點仍未解決時考慮並可能解決這一風險是值得的，而不是等到事情結束後（可能為時已晚）。</p>
</div>

"Big Short" Investor Michael Burry Is Now Betting Against Micron, Nvidia, and Tesla. Should You Be Worried?

- Michael Burry 對人工智能相關股票進行做空，包括 Nvidia 和 Tesla 等。  
- 他認為這些股票價格過於高估，且記憶週期未被市場充分認識。  
- Burry 的做空策略促使投資者審視風險，尤其是 Micron 和 Tesla 的高估值情況。

Motley Fool

卡特彼勒

應用材料

2 倍做空 NVDA ETF - T-Rex

2 倍做空特斯拉 ETF - Tradr

2 倍做多 NVDA ETF - T-Rex

南方兩倍做多特斯拉

1 倍做空 MU ETF - Direxion

2 倍做多特斯拉 ETF - Direxion

2 倍做多英偉達 ETF - GraniteShares

2 倍做多英偉達 ETF - Direxion

1 倍做空英偉達 ETF - Direxion

2 倍做空特斯拉 ETF - GraniteShares

1.5 倍做空英偉達 ETF - Tradr

2 倍做多 TSLA ETF - GraniteShares

2 倍做多 MU ETF - GraniteShares

1 倍做空特斯拉 ETF - Direxion

南方兩倍做空英偉達

Simplify Volt TSLA Revolution ETF

2 倍做多特斯拉 ETF - Leverage Shares

2 倍做空 NVDA ETF - GraniteShares

2 倍做多美光科技 ETF - Direxion

2 倍做多 TSLA ETF - T-Rex

英偉達期權收益策略 ETF - YieldMax

費城交易所 半導體 ETF - iShares

Proshares Ultra TSLA

Leverage Shares 2x Cpd Acclrtd TSLAMnETF

TSLA 周收益 ETF - Roundhill​

Yieldmax Tsla Option Income ETF

2 倍做空特斯拉 ETF - T-Rex

TSLA 期權收益策略 ETF - GraniteShares

xETFs TSLA Daily Income ETF

TSLA 期權收益策略 ETF - YieldMax

Kurv Yield Premium Strategy Tesla TSLA ETF

XI二南特斯-U

2 倍做多 MAG7 指數 ETF - Roundhill

GraniteShares YieldBOOST MU ETF

Defiance Daily Target 2X Short MU ETF

ProShares Ultra NVDA

2 倍做多英偉達 ETF - Leverage Shares

Leverage Shares 2x Cpd Acclrtd NVDAMnETF

NVDA 週配息 ETF - Roundhill​

GraniteShares YieldBOOST NVDA ETF

xETFs NVDA Daily Income ETF

南方兩倍做多英偉達

NVDA 期權收益策略 ETF - YieldMax

XI二南英偉-U

2 倍做多半導體 ETF - ProShares

半導體 3 倍做多 - Direxion

2 倍做空半導體 ETF - ProShares

半導體 3 倍做空 - Direxion

Defiance Daily Target 2X Long AMAT ETF

半導體 ETF - VanEck Vectors

Invesco PHLX Semiconductor ETF

標普半導體 ETF - SPDR

動態半導體 ETF - Invesco

納斯達克半導體 ETF - First Trust

NVIDIA Corporation

《大空頭》投資者邁克爾·伯裏目前正在做空美光科技、英偉達和特斯拉。你需要擔心嗎？