AI’s Capex Tsunami: Why Tech’s Spending Binge Is Remaking the US Economy, Not Bubbling Over

If you want to know what peak cycle feels like, try explaining to your parents that artificial intelligence is now eating 7, 8% of US GDP. No, not the market cap of Nvidia, but the actual capital expenditures of AI-driven tech companies, now outpacing the railroad booms of the 19th century and the dot-com fever of 2000. This isn’t a meme stock rally. It’s a full-blown, balance-sheet-altering, spreadsheet-busting transformation of the American economy. The only thing frothier than the capex budgets are the LinkedIn posts from data center architects.

Let’s get specific: According to Seeking Alpha (Feb 13, 2026), AI infrastructure spending has blown past historical benchmarks. In 1857, railroads were the apex predator of capital formation, soaking up 6% of GDP. The dot-com bubble, for all its irrational exuberance, peaked at about 5%. Now, Big Tech’s AI ambitions are devouring more than 7% of GDP, and the number keeps rising. The narrative that we’re in a bubble is missing the point. This isn’t tulip mania, it’s a capex arms race, and the only losers are the companies that hesitate.

The numbers are staggering. Microsoft, Google, Amazon, and Apple are collectively pouring hundreds of billions into AI chips, data centers, and the power grid upgrades needed to keep the whole thing from melting down. The S&P 500’s tech sector (XLK) is sitting at $139.09, flat on the day, but don’t let the lack of fireworks fool you. Under the surface, the capital intensity is rewriting how investors need to think about growth, margins, and even the definition of ‘tech’ itself. This is not just about cloud GPUs. It’s about who controls the rails of the next industrial revolution.

The macro backdrop is almost comically supportive. Headline CPI just printed at 2.4% for January, down from 2.7% in December (Labor Department, Feb 13), and core inflation is still sticky, but not enough to scare the Fed into a hawkish lurch. Treasury yields are drifting lower (WSJ, Feb 13), giving tech’s duration trade some breathing room. Meanwhile, the market is rotating like a caffeinated squirrel: leveraged growth, precious metals, and crypto have all taken a beating, while the AI narrative keeps sucking up the oxygen. If you’re looking for signs of irrational exuberance, you’ll find them in the options market, not in the actual capex data.

So, is this a bubble? The answer depends on your time horizon and your tolerance for volatility. Yes, valuations are stretched. Yes, the multiples on some AI-adjacent names are pricing in a future where every toaster is sentient. But the real story is that the capital cycle is just getting started. The infrastructure buildout is so massive that it’s creating its own demand, and the feedback loop is only accelerating. If you’re waiting for the crash, you might be waiting a while.

The AI capex boom is also distorting traditional sector correlations. Utilities are suddenly sexy again, because someone has to keep the data centers from going dark. Real estate is quietly repositioning around server farms and edge computing hubs. Even energy markets are feeling the ripple effects, as power-hungry AI clusters drive up demand for electricity at a time when the grid is already creaking under the weight of electrification.

Strykr Watch

For traders, the technicals on XLK are almost boring in their stability. The ETF is parked at $139.09, with support at $137 and resistance at $142. The RSI is hovering around 54, neither overbought nor oversold. The 50-day moving average is flatlining at $138, while the 200-day sits at $135. Volatility is subdued, but don’t mistake calm for safety. The options market is quietly pricing in a pickup in realized volatility over the next quarter, as earnings season collides with the next wave of AI spending announcements.

The real action is in the single names. Nvidia and AMD are the obvious proxies, but the second-order plays, think power utilities, REITs with data center exposure, and even select industrials, are where the smart money is sniffing around. Watch for breakouts above $142 on XLK as a signal that the next leg of the AI capex trade is underway. A break below $137 could trigger a quick flush as the hot money heads for the exits.

The bear case is straightforward: If the Fed gets spooked by sticky core inflation or if a regulatory shock hits Big Tech, the entire capex thesis could unwind in a hurry. But for now, the path of least resistance is up and to the right.

The opportunity set is broad. Long XLK on dips to $137 with a tight stop at $135 looks attractive. For the more adventurous, levered plays on utilities and data center REITs offer asymmetric upside. If you’re feeling brave, shorting the most egregiously valued AI ‘story stocks’ could pay off when the narrative cools. But timing is everything, and the pain trade is still higher.

Strykr Take

This isn’t your grandfather’s tech bubble. The AI capex cycle is real, massive, and only just beginning. Ignore the doomsayers calling for a top. The real risk is being underexposed as the infrastructure buildout reshapes the entire market. Strykr Pulse 72/100. Threat Level 2/5.