The AI investment cycle has become one of the most profitable technology booms in history. Every week seems to bring another report of record chip sales, expanding data center budgets, or larger capital spending plans from the world’s biggest technology companies. Investors have become accustomed to hearing about trillion-dollar opportunities and soaring demand for AI infrastructure.
Yet every dollar poured into one corner of the economy is a dollar that cannot be spent elsewhere. That hidden tradeoff is beginning to show up in corporate earnings, creating an economy of AI haves and have-nots that could shape investment returns for years.
AI’s Biggest Winners Keep Pulling Further Ahead
The contrast became clearer this week. Taiwan Semiconductor Manufacturing (NYSE:TSM | TSM Price Prediction) preannounced second-quarter revenue yesterday that climbed sharply from a year ago, capping another quarter fueled by insatiable AI chip demand. According to the company’s monthly revenue release, June sales reached another record, putting TSM on pace to report another quarter of record earnings when it reports results on Thursday.
That should surprise no one. TSM manufactures the advanced processors used by Nvidia (NASDAQ:NVDA), Advanced Micro Devices (NASDAQ:AMD), Broadcom (NASDAQ:AVGO), Apple (NASDAQ:AAPL), and many of the world’s largest semiconductor companies. As hyperscalers continue spending hundreds of billions of dollars on AI infrastructure, TSM remains one of the biggest beneficiaries.
The same trend is evident throughout the semiconductor ecosystem. Micron Technology (NASDAQ:MU) recently reported record profits as demand for high-bandwidth memory (HBM) continues to outstrip supply, while equipment makers supplying chip fabrication plants have enjoyed years of expanding order books.
These companies occupy the center of the AI spending universe, where nearly every additional dollar of capital expenditure eventually lands.
IBM Shows the Other Side of the AI Boom
Conversely, IBM (NYSE:IBM) offered a glimpse of what happens outside that circle.
The company also just preannounced second-quarter results ahead of earnings, and while management continues highlighting growth in its AI business and the rollout of its new Z17 mainframe, another trend stood out. According to IBM’s letter to shareholders, its Infrastructure segment posted a 7% year-over-year sales decline, while profits also moved lower.
Ironically, IBM isn’t losing because customers have stopped investing in technology. It is because they’re investing almost exclusively in AI.
Companies have finite capital budgets. Servers packed with Nvidia GPUs, networking gear, storage arrays, and new data centers are consuming enormous portions of enterprise spending. That leaves fewer dollars available for upgrading legacy infrastructure, expanding traditional IT projects, or refreshing systems that aren’t viewed as mission-critical for AI deployment.
That resembles the economic lesson French economist Frederic Bastiat illustrated in his famous “broken window” parable. The spending everyone sees on replacing a broken window creates obvious winners — the glazier, who replaces the window — but the opportunities sacrificed elsewhere often remain invisible — the baker, cobbler, the bookseller that the shopkeeper would have spent his money at.
AI infrastructure is producing the same effect: It is creating enormous wealth for semiconductor companies, while quietly delaying investment across other industries.
Investors Should Watch for a Growing Divide
Second-quarter earnings season could become a turning point. If chipmakers, memory suppliers, and equipment manufacturers continue posting record results while companies further removed from AI infrastructure report slowing demand, investors will have fresh evidence that AI’s benefits remain heavily concentrated.
Granted, this imbalance won’t last forever. As AI infrastructure matures and capital spending eventually moderates, money should begin flowing back into software, enterprise systems, and other technology categories. But most Wall Street forecasts suggest that transition remains several years away.
Key Takeaway
In short, investors shouldn’t assume every technology company benefits equally from the AI boom. The biggest winners today remain those selling the picks and shovels — chips, memory, manufacturing capacity, networking equipment, and power infrastructure. Companies operating farther down the technology stack may continue facing headwinds as enterprise budgets remain concentrated on AI buildouts.
Understanding where capital is flowing — and just as importantly, where it isn’t — may prove to be one of the most valuable investing lessons of this AI cycle.
Contact [email protected] for any questions or corrections.
Top Gaining Stocks
Top Losing Stocks