📉 Burry's Top Bet Just Broke

Investors are spending a lot of time trying to figure out which AI company is worth their time and money.

Should they bet on OpenAI? Anthropic? Or one of the many other large language models on the horizon?

But there may be a simpler way to invest in AI: the companies these platforms need to keep growing.

Each large language model requires a massive amount of specialized technology to stay up and running, including memory, storage, networking, semiconductor manufacturing, power, and cooling infrastructure.

MarketBeat’s new The Infrastructure’s Backbone: 10 Stocks Powering the AI Buildout report identifies 10 publicly traded companies that are crucial to the AI infrastructure buildout already underway.

These companies are building the underlying ecosystem that could power the next round of growth for the AI industry.

Get The 10 Names

Michael Burry's largest position just dropped 18% overnight.

Lululemon is roughly 17.4% of his portfolio. So that's not a scratch, that's a chunk of the whole thing.

The trigger: a fiscal Q2 report so bad the stock went straight through $100 in after-hours trading.

Lululemon cut its full-year sales guidance. Again.

Sales fell far more than anyone expected, and the two problems are the ones the company keeps flagging:

Interim co-CEO Meghan Frank called the new outlook prudent and pointed to better product, more marketing spend and tight cost control. Translation: we're spending more money to sell less stuff, and hoping the maths improves.

Here's the part that makes this interesting.

Burry saw it coming. He'd flagged the risk of a weak quarter several times before the print, and he was watching volume rather than the story.

His read: the shareholder base has completely turned over. New holders at these levels are less likely to panic sell than the crowd that came before them.

And he's not walking away. He said he'd buy more if it traded under $100, which is exactly where it went after hours.

He calls Lululemon a "trickster" and says it seems determined to take him somewhere unpleasant. Which is a very poetic way of saying "this thing keeps hurting me and I keep coming back."

Burry's position is down about 20% from his last purchase, and he has a mechanical rule for that: buy more, or sell it. No middle ground.

That's the whole thing. No agonising, no "let's see how Q3 goes." Down 20% forces a decision.

He applied the same honesty to his Tailored Brands position, which he flat out called a mistake. Not a huge one, but the bankruptcy made the loss permanent, and permanent is the word that stings. He said he couldn't take many lessons from it, which is a rare admission from a guy famous for finding lessons everywhere.

Concentration cuts both ways. A 17.4% position falling 18% takes roughly 3% off a portfolio in a single night. That's the price of conviction.

The broader signal is about the athleisure trade. Lululemon spent a decade as the default "premium activewear" answer. Competitors are now taking share in its home market while its growth story in China stalls.

Averaging down works beautifully right up until the moment the business is genuinely broken. The hard part is that both look identical from the outside for about six months.

Is Lululemon a wounded compounder or a melting ice cube? Burry's made his choice. The stock will settle the argument, not the commentary.

Ready to stop scrolling job boards and start building something of your own?

The Hustle's free "Side Hustle Ideas Database" uncovers 100 unconventional, actually-actionable ways to earn on the side.

Inside you'll find:

A curated list of 100 real side hustle ideas spanning e-commerce, services, content, and weird corners of the internet most people never think to look

Opportunities at every level, whether you have 5 hours a week or 50, zero capital or a small runway

Offbeat, under-the-radar ideas you won't find in generic "how to make money" listicles

Subscribe for free today and get instant access. Your next income stream is one click away.

Get The Free Guide

The US economy just added 162,000 jobs when Wall Street penciled in 55,000.

Nearly three times the forecast. And investors responded by selling everything.

Stocks and bonds both dropped on the print. Good news, bad reaction.

August payrolls came in at +162K, against a +55K consensus.

Then the Bureau of Labor Statistics went back and fixed its homework.

That July revision matters most. A month that looked like the labour market cracking is now a month where it quietly grew.

The jobs came from restaurants, bars, and local government education. The information industry shed workers, which is not exactly a shock in the year of "let the AI do it."

Unemployment held at 4.1%, beating the 4.2% consensus and matching July.

7.0 million people are out of work, roughly the population of Hong Kong, and that figure barely moved over the year.

Participation ticked up to 61.6% from 61.4%. That means people who had given up looking started looking again, which is the opposite of what a dying labour market does. Another 5.7 million sit outside the labour force but say they want a job.

Wages did their part too. Average hourly earnings rose 10 cents, or 0.3%, to $37.75, in line with consensus and triple July's 0.1%. Year over year, pay is up 3.1%, above the 3.0% expected and just under June's 3.2%.

Solid jobs, solid wages, stable unemployment. So why did the market flinch?

Strong data means the Fed has less reason to be gentle.

Odds of the Fed holding rates at 3.50%-3.75% slipped to 39.8% from 40.6%. Odds of a 25 basis point hike jumped to 60.2% from 50.4% in a single day.

Not everyone is convinced. David Russell at TradeStation reckons a strong month after two weak ones is more statistical noise than genuine economic acceleration, with unemployment and wages steady underneath.

Michael Kramer of Mott Capital disagrees. He points to the household survey showing real population growth and people re-entering the workforce, and thinks the Fed should hike on September 16.

Chris Zaccarelli at Northlight sees the awkward part. A national election lands less than two months after that meeting, and hiking into an election is not a great look. He counts plenty of reasons to raise rates to fight inflation, and fewer reasons to sit still.

The market has spent months pricing in a friendly Fed. This print makes that harder to justify.

If rate-sensitive corners of your portfolio have been floating on cut expectations, this report is a nudge to check what happens if the cuts never arrive.

Zaccarelli's tell is worth stealing: watch whether stocks shake this off and rally anyway. If they do, the AI build-out and earnings are running the show and the Fed is background noise.

If they don't, rates are back in the driver's seat.

The Dutch central bank just moved 78 tonnes of gold out of New York.

That's $11 billion of metal, roughly the weight of a fully loaded 737, shipped across the Atlantic to sit in London instead.

Central banks don't do this for fun.

DNB says it's strengthening crisis preparedness "in view of increasing geopolitical unrest." No further detail offered.

The stated logic is genuinely dull and genuinely sensible. Bars held at the Bank of England meet international market standards, which per DNB makes them the most easily tradable gold in the world.

Gold sitting in New York? Harder to move in a crisis, the Dutch reckon.

So the scoreboard flipped. Nearly a third of the Dutch reserve now sits in London. New York is down to 19%, a straight reversal of where those two stood before.

DNB never said what "geopolitical unrest" actually means. Very convenient.

The backdrop it declined to mention: 18 months of US-Europe relations going sideways. Tariffs. Trump's threats to seize Greenland. His equivocation over whether America still shows up for European defence.

Every central bank splits its gold across multiple vaults. That part is routine housekeeping.

Choosing which vault you'd least like to be locked out of is a slightly different exercise.

Gold futures rose sharply Thursday, and the Dutch had nothing to do with it.

Fed Governor Christopher Waller said he'd support leaving rates unchanged if inflation pressures keep moderating. That dragged Treasury yields down off multi-year highs and weakened the dollar.

Lower yields plus a softer dollar is gold's favourite weather. It doesn't pay you anything, so it looks a lot better when nothing else does either.

Two dates now own the story: Friday's August non-farm payrolls report, and the Fed's mid-September meeting.

There's a wildcard bolted to the side of it. Renewed fighting in the Middle East has lifted oil four days straight, and pricier oil feeds inflation, which is exactly the argument for the Fed staying put.

So what does any of this mean for your money?

When institutions with 78 tonnes of gold start optimising for "how fast could I sell this," they're preparing for a scenario where that question matters. Whether they're right is a different conversation.

And gold's current rally is running almost entirely on Fed expectations, which one payrolls print can rearrange before lunch.