The S&P 500 and Nasdaq are on track for their steepest monthly decline since March 2025, and if you're wondering what changed, it's simple: Wall Street is finally asking whether AI is worth the money.

According to Reuters, the Nasdaq dropped 0.99% Friday, capping off a brutal month as concerns over Big Tech's AI spending reached critical mass. The Dow fell 1.22%, and the S&P declined 0.66%.

Here's the thing nobody wants to say out loud: AI might be the fiber optic cable boom of the 2020s.

Remember the late 90s? Telecom companies spent hundreds of billions laying fiber optic cable for the internet revolution. The internet did happen. But most of those companies went bankrupt because they overbuilt capacity and couldn't generate returns fast enough to service their debt. The technology was real. The business model was broken.

That's the risk here. Nobody is arguing AI doesn't work. The question is whether Microsoft, Google, Amazon, and Meta are going to make back the $150+ billion they're collectively spending on AI infrastructure this year.

So which stocks are most exposed if AI capital spending slows?

Start with Nvidia. They're the arms dealer of the AI boom. If hyperscalers slow GPU purchases, Nvidia's revenue growth stalls. The stock trades at 50x earnings, which prices in perpetual 30%+ growth. If that slows to 15%, the multiple contracts fast.

Then you have the cloud infrastructure plays - Microsoft Azure, Amazon AWS, Google Cloud. They're building data centers at a record pace, assuming AI workloads will fill them. If enterprise AI adoption is slower than expected, they're stuck with underutilized capacity and massive depreciation charges.

And don't forget the AI software companies that are burning cash to scale. Companies like C3.ai, Palantir, and others are priced for explosive growth. If that growth doesn't materialize, they'll get crushed.

What should investors do? If you have a tech-heavy portfolio, now is the time to ask hard questions. Do you believe AI will drive revenue growth in the next 12-24 months, or is this a 5-10 year buildout? If it's the latter, valuations need to come down.

This doesn't mean AI is fake or won't matter. It means the market might have gotten ahead of itself, and we're now entering the "trough of disillusionment" phase of the hype cycle.

The companies with real use cases and paying customers will survive. The ones burning cash on vague "AI strategies" won't.

Choose accordingly.