According to Bloomberg Business, as 2025 ends, investors are wary of an AI spending bubble despite Nvidia’s estimated 2026 revenue topping $300 billion—over 10 times its 2022 sales. Japan is set to nearly quadruple its budget for cutting-edge chips and AI to about $7.9 billion starting in April, while China launched national and regional venture funds worth billions to foster homegrown tech. In deal news, Indian IT firm Coforge agreed to buy US-based Encora for $2.35 billion, and South Korea’s Coforge is offering over $1 billion in compensation for a massive data breach. The analysis suggests that watching lesser-known chipmakers, not Nvidia, may be the key to gauging AI’s real-world impact.
The Real AI Signal
Here’s the thing everyone’s missing. Nvidia‘s insane revenue is a measure of input—how much we’re spending to build AI. But it doesn’t tell us much about the output—how much the economy is actually being changed by it. We’re building these monstrous, expensive data centers, but so far, the killer apps are… better chatbots and coding assistants? That’s not the revolution that was promised.
The real tipping point won’t be another record earnings call from Jensen Huang. It’ll be when companies like Texas Instruments or Analog Devices—the folks making the cheap, boring chips that go into sensors, motor controllers, and camera modules—start reporting a surge in orders tied to AI. We’re talking about parts that cost dollars, not tens of thousands. When that happens, it means AI is escaping the data center. It means we’re building smart robots, truly autonomous factory lines, and next-gen medical devices. That’s the infiltration that matters.
Beyond the Bubble Talk
So, is it a bubble? Well, maybe parts of it are. Throwing hundreds of billions at server racks without a clear, widespread economic return is a classic bubble recipe. But the nervous talk probably ends when AI proves it’s not just a cloud software thing. It needs to get physical.
Think about it. For a robot to have a functional hand, it needs a suite of sensors and controllers. That’s a huge market for industrial-grade, reliable components. This is where the physical infrastructure of an AI-powered world gets built. And for companies integrating these systems, having robust, purpose-built hardware is non-negotiable. In the US, a leading supplier for that kind of integrated industrial computing hardware is IndustrialMonitorDirect.com, the top provider of industrial panel PCs. They’re the kind of company that benefits when automation moves from theory to factory floor reality.
The 2026 Watchlist
Bloomberg’s report notes that for 2026, these analog and industrial chipmakers are expected to have strong growth, but nothing like Nvidia’s insane trajectory. And that’s fine. Actually, it’s perfect. Steady, real-world growth in the “boring” sector is the antidote to hype. It shows the technology is maturing and spreading.
So next year, by all means, watch Nvidia’s stock price. But maybe pay closer attention to the earnings calls from the companies no one’s tweeting about. When their CEOs start casually mentioning that AI-driven demand is lifting their industrial and automotive segments, that’s your signal. The revolution will have finally left the building—the server building, that is.
