When Goldman Sachs creates a product to help you avoid the hottest trade in the market, that's not a service. That's a signal.

The investment bank just launched an "AI-free" stock index—a basket of companies with zero exposure to artificial intelligence. No chipmakers, no cloud providers, no software companies riding the generative AI wave. Just old-economy businesses doing old-economy things.

Why would one of Wall Street's most powerful firms build a product designed to sit out the AI boom? Because they're not building it for today. They're building it for the day the music stops.

Let's be clear: Goldman isn't calling the top. They're too smart for that. But they are acknowledging something most retail investors don't want to hear—AI stocks are priced for perfection, and when the correction comes, there won't be many places to hide. Unless you've already moved your money somewhere else.

The timing is telling. Nvidia just posted a 63.1% profit margin and guided to $78 billion in quarterly revenue. Microsoft, Google, Meta, and Amazon are spending $700 billion on AI infrastructure in 2026. The Nasdaq is hitting new highs almost daily. And Goldman is quietly rolling out a product that says, "Hey, maybe you want to own something that doesn't depend on this continuing forever."

This is the same playbook they used before the dot-com crash and the 2008 financial crisis. Create a hedge product, market it to institutional clients, and let retail figure it out six months too late. I'm not saying we're headed for a crash. I'm saying Goldman sees enough risk to monetize the other side of the trade.

So what's in an AI-free index? Probably industrials, consumer staples, utilities, healthcare, maybe some financials. Companies that make things, move things, or provide essential services. Boring businesses with stable cash flows and single-digit revenue growth. The kind of stocks that underperformed for the last two years while everyone piled into Nvidia and Microsoft.

The bull case for AI is that we're in the early innings of a decade-long infrastructure buildout. Enterprise adoption is just beginning. The hyperscalers are building capacity ahead of demand, betting that workloads will follow. If they're right, AI stocks have room to run for years.

The bear case is that we're watching a capex bubble inflating in real time. When four companies are spending $700 billion a year on infrastructure, someone has to generate the revenue to justify it. If AI monetization doesn't scale as fast as the buildout, margins will compress and valuations will reset. Fast.

Goldman's AI-free index isn't a call on which scenario plays out. It's a recognition that both scenarios are possible, and some clients want exposure to the second one. That's good risk management. It's also a flashing yellow light.

Here's the part that matters for retail investors. If you own the Nasdaq or the S&P 500, you're already concentrated in AI. The Big Six tech companies—Apple, Amazon, Google, Meta, Microsoft, and Nvidia—account for a massive share of index returns. You don't have diversification. You have six stocks and 494 passengers.

An AI-free portfolio doesn't mean you think AI is going to fail. It means you think AI stocks might take a breather at some point, and when they do, you'd rather own something that isn't correlated to that move. That's not bearish. That's sensible.

Goldman isn't the only one positioning for this. Citron Research just went short on SanDisk, a memory stock that quintupled on AI hype. Hedge funds are quietly trimming exposure to semiconductors. The smart money isn't running for the exits. They're just hedging.

The irony is that if enough institutions start buying AI-free stocks as a hedge, those stocks will outperform purely on rebalancing flows. Value has been left for dead for two years. If capital rotation kicks in, the catch-up trade could be violent. And Goldman will have front-row seats—because they created the product that captured the flows.

Should you buy an AI-free index? That depends on what you own and what you believe. If your portfolio is 80% tech and you're starting to feel uncomfortable, rotating some exposure into non-AI sectors isn't a bad idea. If you're underweight AI and still chasing returns, this product isn't for you.

But here's the real lesson: when the banks start selling insurance, pay attention to what they're insuring against. Goldman isn't worried about AI failing. They're worried about AI stocks getting too crowded. And in markets, crowded trades are the ones that hurt the most when they unwind.