Wall Street’s favorite bedtime story right now is that you can capture the upside of artificial intelligence while neatly insuring yourself against the crash.
It is a lie. Recently making headlines in this space: The Brutal Truth About the Longest Freight Train in History.
Financial pundits love preaching about "smart hedging" and the supposed structural limits of downside protection during a tech supercycle. They tell you to buy out-of-the-money puts on semiconductor ETFs, rotate half your capital into defensive utilities, or hide behind equal-weighted index funds so you don't get burned when the capex bubble pops. They frame this as prudent risk management. In reality, it is a guaranteed mechanism for bleeding out slowly while missing the single largest wealth creation event of our lifetime.
The mainstream narrative—typified by safe, consensus-driven commentary on the limits of hedging the AI boom—gets the math entirely backward. Those commentators assume the primary risk in an AI-driven market is a sudden, catastrophic 2000-style dot-com collapse, and that the only friction in hedging is the slightly elevated cost of options premiums. More information on this are explored by The Economist.
They are fighting the last war. And they are ignoring the brutal mechanics of volatility drag, structural capex shifts, and asymmetric compounding.
The Lazy Consensus Around Downside Protection
Let’s dismantle the standard institutional playbook first. When portfolio managers get nervous about concentration risk in mega-cap tech, they reach for the exact same three tools:
- Index Put Spreads: Buying protective puts on the Nasdaq-100 or semiconductor baskets to cap downside losses.
- Factor Rotation: Selling high-multiple AI infrastructure winners to buy "cheap" value stocks, consumer staples, or regional banks.
- Cash Drag: Sitting in short-term Treasury bills waiting for a 20% pullback that never quite arrives on schedule.
The conventional wisdom argues that while these hedges have "limits"—meaning they aren't 100% airtight—they are necessary guardrails.
Here is what they won't tell you. In a regime shift driven by a foundational general-purpose technology, hedging is not insurance; it is a short position against human productivity. When you pay a recurring premium to protect against a crash, you are introducing a deterministic drag on your compound annual growth rate. Over a five-year horizon, rolling protective puts on high-beta tech equities doesn't just shave a little off the top. It mathematically destroys the right-tail distribution of your returns.
I have watched family offices and mid-sized funds burn tens of millions of dollars over the last three years trying to "collar" their exposure to hyperscalers and chipmakers. Every single quarter, they rolled their upside calls too early, got their shares called away right before an earnings blowout, and watched their downside puts expire worthless. They paid a massive coward’s tax to feel emotionally comfortable. Comfort is the most expensive luxury in finance.
Why Options Markets Are Rigged Against the AI Skeptic
To understand why traditional hedging fails here, you have to look at how implied volatility actually behaves during an infrastructure buildout.
In normal market cycles, volatility spikes when prices fall and compresses when prices rise. Market makers price options based on historical distributions and mean reversion. The AI cycle breaks mean reversion. When a handful of companies are deploying hundreds of billions of dollars in physical capital expenditure—data centers, custom silicon, power grids—the earnings growth isn't speculative vaporware. It is contracted, hard-asset cash flow.
Yet, because retail and institutional traders alike are terrified of a bubble, skew remains chronically elevated. You are paying crash-level premiums for downside protection in an environment where corporate balance sheets are fortress-strong.
Imagine a scenario where you hold a concentrated basket of AI infrastructure leaders. You decide to hedge 100% of your downside beyond a 10% drop using rolling three-month puts. Because implied volatility is priced for an imminent apocalypse, that protection costs you roughly 1.5% to 2% per quarter—or 6% to 8% annually.
If the market trades sideways for two years while earnings catch up to valuations, you haven't stayed flat. You have lost 15% of your purchasing power to options decay alone. Meanwhile, if the sector rips higher by 40%, your net return is severely handicapped, and you still haven't solved the problem of what to do when those puts expire. You are trapped on a treadmill of paying rent to stay invested.
As Nassim Taleb has famously demonstrated throughout his work on fat tails and convexity, buying overpriced insurance in the middle of a crowded panic is a sucker's bet. True antifragility comes from position sizing, not from paying Wall Street dealers exorbitant spreads for synthetic safety nets.
The Fallacy of the "Safe" Sector Rotation
When options become too expensive, the consensus crowd pivots to their second favorite myth: rotating into "AI-insulated" or defensive sectors. They tell you to buy utilities, healthcare, or legacy industrials as a natural hedge against tech valuations.
This is pure delusion. There is no such thing as an AI-insulated sector anymore.
Look at the physical reality of what is happening right now. Utilities are no longer sleepy, bond-proxy dividend payers; they are bottlenecked energy providers scrambling to build gigawatts of baseload power for hyperscale data centers. Industrials are either riding the wave of electrical grid modernization and liquid cooling systems, or they are getting hollowed out by competitors with lower operating costs.
If the AI boom actually collapses—meaning hyperscalers suddenly slash capex and data center construction halts overnight—your "safe" utility stocks and industrial conglomerates are going to crater right alongside Nvidia and Microsoft. Their newfound growth multiples will evaporate instantly. Conversely, if you hide in consumer staples or legacy retail companies that have zero exposure to AI infrastructure, you aren't hedging against a bubble. You are voluntarily buying businesses that are about to get disrupted into obsolescence by deflationary AI software.
Correlation goes to one during a true liquidity crisis, and correlation goes to zero when legacy business models get automated out of existence. Either way, sector rotation fails to protect you.
Answering the Questions Investors Are Afraid to Ask
Can you actually hedge against an AI bubble bursting?
No—not without forfeiting the very returns you entered the market to capture. A true bubble is defined by widespread credit expansion and zero cash-flow backing. While pockets of speculative AI startups will certainly go to zero, the core infrastructure layer is funded by the most profitable monopolies in corporate history. Trying to hedge a bubble that is self-funded by free cash flow is a category error.
What is the biggest risk of hedging tech stocks right now?
Opportunity cost and path dependency. Even if you are right that a 30% correction is coming, if the market rallies 60% before that 30% drop occurs, your hedge will have bled you dry on the way up, and your net entry point will still be worse than if you had simply held unhedged through the storm. Timing both the exit and the re-entry of a hedge requires being right twice. Almost nobody is.
Should you use inverse ETFs to protect an AI-heavy portfolio?
Never. Leveraged and inverse ETFs suffer from daily compounding decay, often called volatility drag. In a choppy, sideways market—which is the exact environment most skeptics predict—an inverse ETF will lose money even if the underlying index ends the period completely flat. Using them as a long-term hedge is financial self-sabotage.
The Only Hedge That Actually Works: Radical Barbell Sizing
If Wall Street’s derivatives and sector rotations are traps, what is the alternative? You stop trying to hedge the asset, and you start structuring your capital.
The institutional obsession with hedging comes from a refusal to use the barbell strategy. Most investors put 100% of their equity allocation into a mushy middle—over-diversified, semi-hedged, heavily benchmarked portfolios that underperform in bull markets and still get crushed in bear markets.
Instead, split your capital into two brutal extremes:
- The Hyper-Aggressive Core (60% to 70%): Own the undisputed apex predators of the AI stack outright. No collars. No covered calls capping your upside. No watered-down equal-weight funds that force you to own the losers alongside the winners. Accept that this bucket will experience 30% to 50% drawdowns. That volatility is the price of admission for 10x compounding.
- The Pure Liquidity Fortress (30% to 40%): Keep the rest in ultra-short-duration sovereign debt or pure cash equivalents. Zero credit risk. Zero equity correlation. Zero options decay.
Think about the mechanics of this setup compared to a hedged portfolio.
When the AI sector rips higher by 50%, your unhedged 65% equity sleeve captures every single basis point of that upside, easily outperforming a 100% equity portfolio weighed down by options premiums and defensive value traps. And when the inevitable 40% market panic hits? You don't have to pray that your counterparty honors an over-the-counter derivative or wonder when to sell your protective puts. You have a massive war chest of liquid cash sitting ready to buy distressed apex assets at fifty cents on the dollar.
You aren't paying anyone a fee to hold cash. In fact, you are getting paid yield to wait.
The Brutal Truth About This Approach
I won't pretend the barbell strategy is painless. It comes with a massive psychological flaw that breaks most investors: you will look like an idiot in both directions for months at a time.
During a euphoric melt-up, your cash pile will feel like dead weight, and you will be tempted to deploy it right at the top. During a brutal tech sell-off, watching your unhedged core drop by a third without a put option cushioning the blow requires a stomach of iron. Most people lack the discipline to sit still while their net worth swings wildly, which is precisely why Wall Street makes billions selling them illusionary hedges.
If you cannot stomach a 35% drawdown on your risk capital, you shouldn't be in AI equities at all. Buy Treasury bonds and go play golf.
Stop Insuring Your Winners and Start Shorting the Victims
If you still insist on running an active hedge book—if you run a long-short mandate or simply refuse to hold cash—stop shorting the AI winners. Stop buying puts on the companies building the future.
Hedge the AI boom by shorting the companies AI is going to destroy.
The market is obsessively focused on who wins the AI race, leaving massive mispricings among the legacy incumbents sitting directly in the blast radius. Look at bloated enterprise software companies charging per-seat licenses for basic workflow tools that an AI agent can replicate for pennies. Look at legacy business process outsourcers, low-tier digital ad agencies, and mid-market consulting firms whose entire billing model relies on armies of junior analysts doing repetitive cognitive labor.
If the AI boom accelerates, those legacy incumbents get decimated—giving your short book massive, asymmetric downside capture. If the AI boom stalls out and enterprise spending freezes, those same bloated software and consulting firms still get crushed because their customers cut discretionary IT budgets first.
That is a real hedge. It pays you when the thesis works, and it protects you when the macro environment sours.
Stop paying Wall Street a tax to protect you from upside volatility. Take off the training wheels, size your positions so you can survive a crash without flinching, and let compounding do its job.