What just happened in the markets might go down as the most misunderstood week of 2025.

Big Tech just posted one of the strongest rounds of earnings in history … and yet, half of the sector tanked.

Meta (META) and Microsoft (MSFT) sold off.

Amazon (AMZN) and Alphabet (GOOGL) surged.

Apple (AAPL) barely moved.

Same economy. Same quarter. Same AI boom. So why did investors react like they were watching five different movies?

That’s the question we tackle in this week’s Being Exponential.

Because beneath the headlines lies a story that could determine the next phase of the bull market… or mark the beginning of its unraveling.

Yet, here’s where things get truly interesting.

While Wall Street frets about short-term noise, a new wave of capital spending is building beneath the surface…

We’re talking about the unsung backbone of the AI revolution:

• Rare earth miners powering data centers.
  
• Nuclear innovators keeping the grid alive.
  
• Storage companies enabling round-the-clock AI processing.

And that’s where the smart money is quietly positioning itself right now.Be sure to watch the latest Being Exponential With Luke Lango to learn more. Just click the video below:

Follow the Money in AI Stocks: The $380 Billion AI Capex Funnel

If you’re wondering how long this boom can last, one key is to follow Big Tech’s money trail. 

In 2024, the four AI superpowers (Alphabet, Amazon, Meta, Microsoft) spent over $200 billion on capital expenditures (much of it AI-related). In 2025, that figure is expected to hit ~$380 billion, and then over $500 billion in 2026 – a 34% jump next year alone. These are historic levels of investment, aimed at one thing: meeting the red-hot demand for AI. As long as that torrent of cash flows in, it will percolate throughout the entire AI ecosystem, creating opportunities up and down the supply chain.

Think of the AI supply chain as a modern-day gold rush, and Big Tech is supplying the dynamite. Here’s where their billions are flowing – and which industries could be the big winners:

• Chips and Computing Power: Every AI model needs heavy-duty silicon. 2023’s AI frenzy made Nvidia (NVDA) a $1-plus trillion company, and its GPUs are still selling faster than it can make them. Demand for AI chips is so high that major cloud providers are building custom chips and still buying all the Nvidia GPUs they can get.
  
• Memory and Storage: Training and running AI models gobble up huge amounts of data – boosting demand for memory chips (DRAM) and high-speed storage. In fact, memory prices are rising again (Samsung and SK Hynix recently hiked prices) as AI data center orders pour in. That’s a boon for suppliers like Micron Technology (MU) and Western Digital (WDC), whose stock just spiked after a strong quarter.
  
• Networking & Cloud Infrastructure: All those AI chips need to be connected in sprawling data centers, and companies like Arista Networks (ANET) are benefiting as cloud giants upgrade their networks to 400G/800G speeds for AI workloads. Cisco (CSCO), Juniper (JNPR), and specialized optical networking firms like Lumentum (LITE) and Ciena (CIEN) are likewise getting a piece of the action, providing the fiber-optics, routers, and switches that link AI supercomputers together. Arista’s CEO said AI demand is a “once-in-a-generation” networking upgrade cycle – and their recent earnings back that up.
  
• Data Center Builders & Power Suppliers: AI eats enormous amounts of electricity. The biggest bottleneck in building more AI capacity isn’t chips; it’s power. (As Andy Jassy quipped, “maybe the bottleneck is power” in data centers right now.) This means companies that can deliver power solutions are in high demand. For example, Bloom Energy (BE) – which makes fuel cells for on-site power – reports a surge in orders from data center operators who need reliable, round-the-clock juice for AI rigs. Nuclear energy is also staging a comeback thanks to AI: firms like BWX Technologies (BWXT), which makes small reactor components, say demand for micro-reactors is “unprecedented” as the industry explores mini nuclear plants to power server farms.
  
• “Picks & Shovels” Equipment: In any gold rush, the tool suppliers often win big. In AI, that means the makers of semiconductor manufacturing equipment (think Applied Materials (AMAT) and Lam Research (LCRX) for chip fabrication tools). It also means testers like Teradyne (TER), which just had a blowout quarter by selling test gear for AI chips. As long as Big Tech keeps spending on AI, these suppliers will keep seeing huge orders. And Big Tech has no intention of tapping the brakes yet.

All told, this AI CapEx super-cycle is creating a rising tide for a wide range of stocks. 

The recent market dip has knocked many of these “picks and shovels” stocks down from their highs – which, in my view, presents an attractive buy-the-dip opportunity. 

Not an AI Bubble (Yet) – But Know What to Watch

With all this euphoria, prudent investors still ask: How will we know when the AI boom is truly over? No boom lasts forever, and there will be an “AI reckoning” eventually. While I agree, I don’t see it in the immediate future. In our podcast, we outlined some yellow flags to monitor going forward, even as you ride this wave:

• Frothy IPOs and M&A: Early signs of bubble behavior often include a rush of IPOs (initial public offerings) and frenzied corporate deal-making. We’re seeing some of that now – nearly 300 IPOs in 2025, up 50% from last year, and a spate of big mergers in tech/AI lately.
  
• Economic K-Shaped Divide: The AI boom is creating a K-shaped economy, where the “upper leg” (wealthy consumers, big firms) is thriving, but the “lower leg” (middle and working class, smaller firms) is struggling. For example, luxury gym operator Life Time Group (LTH) saw memberships and prices jump with no pushback. Meanwhile, budget eateries like Shake Shack (SHAK) and Chipotle (CMG) report customers cutting back as prices bite. If AI-driven prosperity keeps concentrating at the top, it could fuel a social/political backlash.
  
• The Fed’s New Dilemma: We’d be remiss not to mention macroeconomics. The Federal Reserve has been aggressively raising interest rates to cool inflation – normally a big headwind for growth stocks. But intriguingly, the Fed might be losing its grip on this market cycle. Why? Because AI is an exogenous force boosting productivity and growth in ways Powell and team can’t easily model.
  

So far, none of these yellow flags are red lights just yet. The data says the AI engine is still revving: enterprise AI spending is accelerating, not slowing, and consumers are still captivated by AI-enhanced products. 

In summary, the AI boom is rewriting the rules … for now. Tech titans are defying gravity, and a broad swath of “pick-and-shovel” companies are thriving in their slipstream. If you stay informed and selective, you can ride this exponential trend while managing the risks. 

Don’t let unfounded bubble fears scare you away from what could be the defining investment story of the decade. As always, do your homework, position size wisely, and keep an eye on the data.

For deeper insights and specific stock ideas, check out the full discussion with on our latest Being Exponential podcast. We unpack dozens of AI stock picks – from rare earth miners to software standouts – and map out exactly how to play the next leg of this AI hypergrowth cycle. It’s a must-listen for anyone looking to navigate the opportunities (and eventual pitfalls) of the AI era