© 2026 Monument Traders Alliance, LLC
Ken Griffin says AI is almost certainly a bubble.
The Citadel founder said it yesterday at his firm’s markets conference in New York. He said any radical change to the economy usually comes wrapped in a bubble, and the companies that survive it can lead for decades. He told the crowd he wishes he’d kept his Amazon (AMZN) stock from the dot-com days.
The ones that don’t survive can take a lot of portfolios down with them.
You may be thinking you don’t own any AI stocks, so this isn’t your problem. But if you own an S&P 500 index fund, you do.
By most counts, AI-related companies now make up somewhere around 40% of the S&P 500. Every dollar you put in that fund, about 40 cents rides on the AI trade.
So what happens if it doesn’t work out?
On October 16, the Nasdaq fell more than 2% as investors started asking hard questions about the debt behind the AI boom. If this trade breaks, last week is going to look like a walk in the park.
Nassim Taleb made the term “black swan” famous in his 2007 book. A black swan is a rare event nobody saw coming that changes everything when it hits, like 2008 or the COVID crash in 2020.
Most market models assume returns fall on a neat bell curve, where giant moves almost never happen. Real markets have fat tails, which means those crashes show up a lot more often than the math predicts.
Tail-risk hedging means owning something small that pays off BIG when the market crashes. When the rest of your portfolio drops 30%, the hedge shoots up and covers a chunk of the damage.
It works like insurance on your house. You pay a premium every year, you pick a deductible, and hope you never file a claim.
The classic tail trade is a deep out-of-the-money put. A put option gives you the right to sell at a set price, so it gains value when the market falls.
Out-of-the-money means the strike sits below where the market trades today, and deep out-of-the-money means way below, 15% or 20% or more. The market almost never falls that far in a few months, so those puts are cheap.
Most of the time, they expire worthless. But when a black swan hits, a put that cost pennies can be worth many times what you paid, which is how Universa, a fund built around this trade, told clients it returned 3,612% in March 2020.
If your worry is AI, you can buy those puts on the Nasdaq-100 instead of the S&P 500. That’s where the AI names carry the most weight, so it tends to fall harder when they do.
Broader hedges can be smart too. A put spread, where you buy one put and sell a cheaper one below it, cuts your cost but caps your payout. Calls on the VIX, Wall Street’s fear gauge, gain when fear spikes, and the VIX hit a record close of 82.69 in March 2020.
Tail traders keep it small. A sliver of the portfolio goes to protection, and they roll it forward, buying new puts a few months out as the old ones expire.
In a calm market, those puts expire worthless over and over, and you keep buying the next batch. Verdad Capital estimates a mix of hedges costs about 2% a year over the long run.
That cost is why people quit right before the fire. CalPERS, the biggest public pension fund in America, dropped its crash hedge by January 2020 because it cost too much. Two months later, that position would’ve paid out more than $1 billion.
Goldman Sachs Asset Management found a tail hedge earns next to nothing on its own. Because it caps your losses in a bear market, it lets you keep more money in growth stocks the rest of the time.
So a hedge that loses money most years is doing its job. You pay that cost so you can stay in the AI trade without losing sleep over it.
Whatever you use, decide what you’re willing to spend on protection each year before you buy it, and write that number down.
When the market’s ripping and the hedge has bled for a year, that number is what keeps you from doing what CalPERS did.
P.S. Last Friday’s drop pushed the VIX right up to 20. That’s the level CJ calls critical, and he’s going to tell you what he thinks happens next on MTA LIVE tomorrow at 2 p.m. ET. He and Nate Bear are also getting into next week’s earnings from Microsoft, Amazon, and Google (a big chunk of that 40% I was talking about), and the sectors money’s been rotating into.
