원문 보기
When Michael Berry announced a billion-dollar bet against Nvidia and Palantir, Palantir's CEO appeared on CNBC and called it complete madness. Berry responded and did something even stranger. He closed his fund so that no one could track his bets anymore. This was the beginning of this crazy story. Four days ago, the story became even more bizarre and took a turn that almost no one expected.
By the end of this video, you'll learn what happened and why some analysts believe Berry may be right after all—a prediction that, if confirmed, would have serious implications for the U.S. economy and the future of the stock market. For those who don't know, Michael Berry is the same guy from the movie "The Downside," a hedge fund manager who bet against the real estate market in 2008, when no one believed it would crash, and made a fortune doing it.
In 2021, he called the market the biggest speculative bubble in history, and now he's making a similar prediction again. only this time it's not houses, it's AI. Here's what he actually did. About 80% of his current portfolio now consists of put options against Palantir and Nvidia. A put option is essentially a bet that the stock will fall in price.
You choose a price, and if the stock falls below it, you receive a payout similar to an insurance payout. In total, his bet covers about $1.1 billion worth of Nvidia and Palantir shares. As for Nvidia, he chose the number 110 as his target. If Nvidia shares fall below this level by 2027, he will make a big profit, and his bet on Palantir is set even more aggressively.
So why does he think it will work ? Well , it boils down to three main statements. First, he says that Nvidia's huge sales are not driven by many of the regular companies that buy chips. This is a small number of giant customers , which means that the growth looks broader and healthier than it actually is, once you put aside the financial jargon.
Second, and this takes time to understand: he accuses big cloud companies of playing hardball with how they report AI hardware costs. Here is a simple version. If a company buys a truck and says it will last 20 years, it only writes off a small portion of that cost from profit each year. But if the truck breaks down after 5 years, it turns out they have been secretly overstating their profits every year.
Michael Bury actually argues that cloud giants are doing just that with their AI servers, claiming they will last much longer than they actually will, making today's profits appear much larger than they actually are. He even names a specific figure: up to 176 billion in hidden costs by 2028. And third, his final statement: the hundreds of billions of dollars in contracts that these companies signed for data centers are not yet legally required to be reflected in their reporting.
Think of it like a credit card bill that doesn't get posted until you get your statement, except it's hundreds of billions of dollars of liability that remain in plain sight. Bury didn't make this up. He cites a report from the Bank for International Settlements, essentially the organization that monitors the global banking system , and they pointed to this very risk.
So, immediately after disclosing these bets and exchanging “shots ” with the CEO of Palantir, Bury did something that no hedge fund manager almost ever does. He withdrew his fund from SEC registration. This means that he is no longer legally obligated to inform the public about his actions. He returned the money to investors and is probably now using his own funds.
Before that, he wrote to them: "My assessment of the value of the securities does not now coincide, and has not coincided for a long time , with the market." He then posted a screenshot of his funds with the status “suspended” and the caption “ Moving on to something better.” He called his new project " Cassandra Unleashed" - in honor of the myth of the prophetess who was always right, but whom no one believed.
In the near future, whatever we learn about his bets, we will only learn because he himself chooses to tell us. So, the big question everyone is asking is simple. Does it work? This week—no, absolutely not. For example, Palantir reported its best quarter ever. Revenue grew by 93% year-on-year. Their CEO called it " otherworldly." The stock jumped almost 30 % in one day.
Although they were down 40% before the report, this single report changed the future for Palantir. If Bury's position had previously brought in money, a significant portion of it had evaporated in about 24 hours . But betting against a company is one thing, but betting against an entire industry that just posted such numbers is either true confidence or a real mistake.
It's hard to say right now which direction things will go. Part of the reason is that Bury is not just betting against Palantir. Everything is much deeper. It's about how money moves in this whole industry. Nvidia just invested up to 100 billion in OpenAI. OpenAI instead enters into a $300 billion cloud deal with Oracle—the largest cloud contract in history.
Oracle then uses this deal to justify purchasing tens of billions more worth of chips from Nvidia . Watch your money—it's a vicious circle. The cash leaves Nvidia's balance sheet as an investment and comes back as revenue after the round through the other two companies. Analysts now estimate the total value of these cyclical deals at around $800 billion, with OpenAI, at the heart of it, projected to lose around $13 billion this year alone.
That's why Berry's bet is n't really on one stock falling. It's about a part of this industry perhaps demonstrating its own growth by selling to itself, rather than to the outside world. And this is not a theory, you can check it yourself. Two weeks ago, Nvidia shares fell 4.5% a day after news that it was considering underwriting a quarter of a trillion dollars in funding for a single OpenAI data center.
And this alleged bubble is not limited to technology companies. Just a few weeks ago, before all this, Berry did something no one expected— he opened a short position in Caterpillar. This is not a chip or software company . It is a giant in the construction and mining industries. What is the connection?
Well, AI data centers require a huge amount of energy, and Caterpillar is positioning itself as an infrastructure provider for AI. Only one deal covers 2 gigawatts of natural gas generators for hyperscale data centers. Investors began to value the company like any other winner in the AI era. Berry is betting that this trade is as overvalued as the chips themselves, and it's further evidence that he believes the problem is much broader than a few companies.
He is playing against the entire sector. And don't get me wrong: just a few weeks ago, it seemed like Berry had predicted everything, with Nvidia down about 20% from its May peaks. Oracle has fallen 62% from its last high. Palantir— about 43%. So for a moment, it seemed like everything was going according to his plan.
But the recent resurgence, plus earnings that consistently beat expectations, have effectively shattered his theory for now. The semiconductor ETF, SMH, is up 54% year-to-date. These numbers are hard to ignore, especially considering how unsuccessful some of his recent predictions have been. Berry called the market overvalued back in 2023, and since then it has grown by about 131%.
So, many people are skeptical about his latest moves, despite the fact that he backs up his words with real money. So what does this bring us to? Well, Berry has made similar predictions before and was absolutely right. At the same time, he was sometimes too hasty and made mistakes, especially recently.
Many of his arguments, particularly regarding depreciation and the cyclical funding mechanisms supporting AI, are certainly worth exploring. But whether they are enough to cause the AI “bubble” to burst is another story. If he turns out to be right, there is no doubt that the market and the economy as a whole will suffer seriously.
AI companies and the hype surrounding them have driven much of the market growth over the past two years, and the end of this phase could be a 1999-style moment for Wall Street . Thank you for watching. Please click the " like" button and subscribe if you enjoyed the video. I noticed that during the last upload, about 89% of viewers were not subscribed.
If you like the content and would like to see my future videos, please take a moment to click this button.
Comments 0