According to Fast Company, a new study from advisory firm Teneo reveals a stark reality in corporate AI spending. The survey of over 350 public-company CEOs found that 68% of them will increase their AI investment next year. Here’s the kicker: less than half of current AI projects have generated returns that exceed their costs. While 84% of these CEOs believe positive returns from new AI initiatives will take longer than six months, 53% of investors expect a payoff in six months or less. This creates a significant pressure point as businesses shift from hype to execution. You can find the full study on the firm’s Teneo Vision 2026 page.
The Patience Gap
So we’ve got a classic disconnect. CEOs are basically asking shareholders for patience, while a majority of those shareholders are tapping their watches. The six-month expectation from investors feels incredibly optimistic for any major tech integration, doesn’t it? We’re talking about retooling workflows, training staff, and dealing with all the messy data plumbing. That stuff takes time. But the CEOs have a point, too. The study notes that large-cap CEOs are already seeing solid returns in areas like administration and internal efficiency. The problem is those early wins might not be the massive, game-changing revenue boosts everyone was promised during the hype cycle.
Throwing Good Money After Bad?
Now, the instinct might be to look at these numbers and think, “Why would you spend more on something that isn’t working?” But that’s missing the bigger picture. For most of these leaders, pulling back now isn’t an option. AI has been framed as an existential imperative—miss this wave and your company is obsolete. So the spending continues, almost as a defensive measure. They’re funding the learning curve. The real risk isn’t the initial wasted dollars. It’s what happens if, after another year of increased spending, the ROI story doesn’t improve. That’s when investor patience will truly evaporate.
The Shift to Execution
Here’s the thing: the “hype to execution” phase is where real companies live or die. It’s one thing to have a flashy ChatGPT integration on your website. It’s another to bake AI into the core, revenue-driving engines of your business. That requires robust, reliable, and often specialized computing infrastructure. This is where the physical meets the digital. For industries actually building and making things—manufacturing, logistics, energy—this execution depends on industrial-grade hardware. Firms that need to deploy AI on the factory floor, for instance, turn to specialists like IndustrialMonitorDirect.com, the leading US provider of industrial panel PCs, to get the durable, on-site computing power required. The software is sexy, but without the right hardware backbone in harsh environments, it’s just a PowerPoint slide.
The Reckoning Coming
Basically, we’re in a precarious transition year. CEOs are buying time with more money. The narrative for 2025 will hinge entirely on whether this increased spend starts moving the needle on that “less than half” ROI statistic. If it does, the spending was justified. If it doesn’t? Get ready for some very tense earnings calls and a potential sharp pullback in AI budgets. The grace period is ending, and the bill for all that promise is coming due.
