A financial research firm's economist has put a number on something many business owners have quietly suspected: for most companies outside the tech sector, AI spending isn't showing up on the bottom line. The finding doesn't mean AI tools don't work. It means the money going into them isn't translating into measurable profit for most of the businesses buying them.
The analysis, produced by an economist at Apollo Global Management, looked at corporate investment in AI tools and systems across non-tech industries and compared it against financial performance. The conclusion: outside of the technology sector itself, the payoff has been effectively zero. Companies are spending on AI subscriptions, custom tools, and pilot projects, but that spending isn't consistently showing up as higher revenue, lower costs, or better margins.
This isn't the first data point suggesting a gap between AI hype and AI results. Surveys from consulting firms over the past two years have repeatedly found that a majority of corporate AI pilots never make it past the testing phase, and that even among companies that deploy AI tools at scale, most can't point to a clear return on investment. What's different here is the framing: this is coming from a Wall Street economist looking at the issue through a financial lens, not a tech research firm surveying executives about their intentions.
The tech sector itself remains the exception. Companies that build or sell AI tools, chips, or cloud infrastructure have seen real revenue growth tied directly to AI demand. The disconnect is specifically about the companies buying AI tools to use internally — retailers, manufacturers, professional services firms, and similar businesses — rather than the companies selling them.
Why it matters
This finding fits a pattern familiar from past technology cycles: the sellers of a new technology tend to profit first and fastest, while the buyers take longer to figure out how to use it productively, if they ever do. Something similar played out with early cloud computing and, before that, with the first wave of corporate website and e-commerce spending in the late 1990s. The tools eventually paid off for many businesses, but not on the timeline the early marketing promised.
What this means for small businesses
If you've adopted an AI tool and haven't seen a clear return, you're not behind the curve — you're in the majority, according to this data. That's worth remembering the next time a vendor pitch implies that everyone else is already profiting and you're falling behind by not signing up.
It also argues for tighter tracking of what you're actually spending on AI tools and what you're getting back. A subscription that saves an employee two hours a week is a measurable win. A tool that sounded promising in a demo but nobody on staff actually uses regularly is a cost with no offsetting benefit, and those add up quietly on monthly statements.
None of this means AI tools are useless for small businesses. Plenty of narrow applications — drafting routine correspondence, summarizing documents, handling basic customer service queries — produce real time savings. The gap identified in this report is at the level of broad corporate strategy and large-scale deployment, not necessarily at the level of a five-person shop using a chatbot to answer FAQs.
What to watch
Watch for how vendors respond to data like this. Expect more emphasis on case studies and ROI calculators in sales pitches, and watch whether major software providers start bundling AI features into existing subscriptions rather than charging separately — a sign they may be having the same trouble proving standalone value.
The bottom line
Before renewing or adding any AI tool subscription, audit whether it's producing a measurable result — time saved, revenue gained, or costs cut — and drop the ones that aren't, regardless of how promising they sounded at signup.