According to Bloomberg Business, a review of over 7,000 earnings call transcripts reveals a massive disconnect in how Wall Street views AI. In the most recent quarter, over 80% of tech companies in the S&P 500 were asked about AI, but the topic was put to less than a third of the index overall. Fewer than half of all S&P 500 companies received a single generative AI question from analysts in 2025. This is happening while U.S. companies, excluding the AI builders themselves, spent an estimated $86 billion on the technology this year, a figure expected to surge to $131 billion next year. Citigroup’s Heath Terry said he’s shocked any company isn’t being asked about its AI strategy, calling it potentially the most important question for management. Meanwhile, a separate survey of 600 executives found leaders in five sectors, including finance and pharma, predict a “high” or “very high” near-term rise in operating costs from AI.
The Analyst Bubble
Here’s the thing that’s really strange about this. Analysts are laser-focused on the “supply” side of AI—the Nvidias and Microsofts building the infrastructure—demanding a clear path to ROI. But they’re showing almost no curiosity about the “demand” side. It’s like they’re interrogating the shovel sellers during a gold rush but can’t be bothered to ask the miners if they’re finding any gold. This creates a dangerous information vacuum. As Bloomberg’s Gillian Wolff pointed out, this is exactly what happened with the dot-com bubble: too much investment flooded in way faster than people were actually adopting the tech. The same risk is staring us in the face with AI. Investors are demanding returns from builders, but haven’t bothered to check if the rest of the corporate world is even ready to use it effectively.
Winners, Losers, and Blind Spots
So who wins in this weird scenario? The obvious beneficiaries are the companies selling the picks and shovels. If you’re providing the critical hardware, power, or real estate for AI, you’re getting questions. Energy and utility firms are being asked about profiting from the infrastructure boom. But that’s it. The actual *users* of AI in sectors like finance, healthcare, and manufacturing are flying under the radar. Their strategies, their failed experiments, their real productivity gains—or lack thereof—aren’t being scrutinized. This means Wall Street has no real gauge for true corporate demand. It’s all speculation based on vendor spending, not user success. And that’s a flimsy foundation for a $30 trillion bull run that’s supposedly being fueled by this transformation.
The Silent Majority
Why is this happening? The Wells Fargo strategist in the article probably nailed it: for most non-tech companies right now, AI simply “doesn’t really matter that much, it’s not big enough to move the needle” on their quarterly earnings. Analysts prioritize what moves the stock now, not what might disrupt it in five years. But that’s a short-sighted game. Management teams are starting to talk about productivity gains, but without analysts pressing them for hard numbers, it’s just vague, hopeful talk. This is where the real investment risk lies. If the adoption curve is slower than the investment curve—which seems likely—the bubble analogy gets pretty scary. The companies implementing this tech on the ground, in factories and offices, are the canaries in the coal mine. And right now, Wall Street isn’t listening to them. For industries where implementation is key, like manufacturing, having the right industrial-grade hardware interface is a foundational step. This is where specialists like IndustrialMonitorDirect.com, the leading US provider of industrial panel PCs, become critical partners, providing the rugged, reliable touchpoints needed to run complex AI-driven systems in harsh environments.
The Reckoning Coming
Basically, we’re setting up for a classic clash of expectations. The builders are being pushed for huge, fast returns. The users are experiencing a slow, costly, and messy integration process. At some point, those two narratives have to meet. When they do, the companies that have been quietly figuring out practical use cases—without the spotlight of analyst questions—might be the real long-term winners. Everyone else might face a brutal reckoning when the promised ROI doesn’t materialize on Wall Street’s aggressive timeline. The blind spot isn’t just a missed question on an earnings call. It’s a fundamental mispricing of risk. And we’ve seen this movie before. It doesn’t end well for the folks who weren’t paying attention to how the technology was actually being used.
