CONTEXT
Business.com and Dialog published their second annual AI adoption survey in January 2026. The study covered 1,009 workers at U.S. companies with fewer than 250 employees. It is one of the more careful looks at what is actually happening inside small businesses on AI, not what owners are hoping is happening.
The headline finding most coverage focused on was adoption growth. Investment in AI among small businesses has risen from 36% in 2023 to 57% in 2025. Usage is up. Daily use is up. Deployment across business functions is up.
Those numbers look like progress. One finding buried deeper in the study tells a different story.
Thirty percent of small business employees say they act more optimistic about AI among colleagues than they actually feel.
That is not a rounding error. It is nearly one in three people on your team performing enthusiasm they do not have.
STRUCTURAL ANALYSIS
This kind of gap does not appear randomly. It forms when the culture around a topic makes skepticism feel risky.
Most small business owners have signaled, directly or indirectly, that AI adoption is a priority. They have purchased tools, attended demos, or talked about the need to keep up. The message to the team, whether intended or not, is that AI is the direction and resistance is the wrong posture.
When employees sense that, they adjust. Not dishonestly, but practically. They tell you the tool works. They mention it in meetings. They do not bring up the friction, the confusion, or the suspicion that the problem was not what the tool was supposed to fix.
The survey also found that 45% of small business workers worry that adopting too much AI could harm their company’s reputation. That concern is not being said out loud. It is being managed quietly, on the employee’s side, out of your view.
The result is a specific kind of information failure. You are making decisions about AI adoption based on what your team reports, and your team is reporting what they believe you want to hear. The gap between those two things is where implementation problems form, where tools get abandoned, and where owners eventually conclude that AI did not work for their business.
ORDER CHECK
Before you draw any conclusions about how AI is working in your operation, four questions need honest answers. Not survey answers. Not meeting answers. Documented answers.
Have you created conditions where your team can tell you something is not working without it reflecting badly on them? If skepticism reads as lack of effort or resistance to change, you will not hear it.
Do you know which tools your team is actually using versus the tools they report using? Usage data and self-reported usage are not the same thing.
Have you asked anyone on your team, directly and privately, what they actually think is and is not working? Not in a group setting. One on one.
If your team stopped using your current AI tools tomorrow, would you know within a week? If the answer is no, the tools are not load-bearing yet. Optional tools do not produce measurable outcomes.
If you cannot answer these cleanly, you do not have a clear picture of where you actually stand.
DECISION
The adoption numbers are real. The enthusiasm gap is also real. Both things are true at the same time.
The owners who will get the most from AI are not the ones who move fastest. They are the ones who know what is actually happening inside their operation before they decide what to do next.
That requires a different kind of conversation with your team. Not a pep talk about AI’s potential. A direct question about what is working, what is not, and what they have not said yet because they were not sure it was safe to say.
The tools are not the problem here. The information environment is. Fix that first.
If you cannot answer those four questions cleanly, you are making decisions without a clear picture. The TAKTOS workshop is designed to help you see what is actually happening inside your business before you move. https://taktos.ai/workshops/
Decide well,
—Chuck