When the airlines replaced ticket agents with self-service kiosks, I thought it was a terrible idea. I wanted a person who could rebook me when my connection fell apart at 10 p.m. in Dallas. But those kiosks gave me more transparency, more control, and faster results than any single agent ever did. The airlines solved a customer problem, and I didn’t just accept the technology. I preferred it.
Now think about retail self-checkout: The same concept, but with a completely different outcome. Retailers solved a company problem, cut their labor costs, and pushed the work onto customers who got nothing in return. The technology worked fine. Acceptance never materialized because no one gave customers a reason to care.
THE SUCCESS FORMULA
I have spent over 30 years helping organizations improve, and that contrast sits at the center of a formula I return to constantly: S = Q x A. Success equals the quality of the solution multiplied by the level of acceptance. Simple multiplication, with a brutal consequence. If acceptance is zero, the whole equation is zero.
Right now, we are watching the global economy pour trillions of dollars into the Q side of this equation for artificial intelligence. Better models. Faster chips. Agents that can draft your contracts and analyze your supply chain. The pace of change and improvement is overwhelming. On the technology side, we are swimming in quality.
But acceptance is cratering. And this is not a technology gap. This is a leadership problem, and almost nobody wants to talk about it.
WE HAVE BEEN HERE BEFORE
I have seen this pattern destroy improvement initiatives for three decades. Companies fall in love with a solution and forget to bring their people along. They pour everything into the “what” and almost nothing into the “why”. With AI, though, we are doing something worse than neglecting acceptance. We are actively poisoning it.
I lived through a version of this myself. When I was in college, the internet was just emerging. It existed, but my professors would not accept any source that wasn’t physically in the campus library. So instead of learning how to use this brand-new technology that was about to reshape every industry on the planet, I paid the university to teach me the Dewey Decimal System and to walk the stacks. Not a skill I have used since graduation. The university didn’t ban the internet because it was unreliable. They banned it because the people in charge didn’t understand it yet. And when you don’t understand something, the easiest response is to call it cheating.
We are repeating that mistake with AI at a scale that should concern every business leader. More than half of students are afraid to use AI for legitimate academic support because they fear being accused of cheating. An educator quoted in Fast Company said that anything you send home with students, you must assume is “being AI’ed,” as if using the most powerful productivity tool of our generation is inherently dishonest. We are conditioning the next generation of workers to treat AI as something to hide, something that makes their work less legitimate. Even this publication prohibits the use of AI in article submissions. I understand the intent, but the message it reinforces is the same one eroding acceptance everywhere else.
The workplace tells the same story. A Udacity study found that one in three workers don’t feel comfortable admitting they use AI on the job, and 32% are using unauthorized AI tools because their employers won’t provide approved ones. That creates exactly the kind of security and compliance problems that leadership was worried about in the first place.
The fix requires leaders to care as much about acceptance as they do about the technology. The companies getting AI right have one thing in common: their senior leaders actively model AI use and drive adoption from the top. They don’t just buy the tools; they buy utility and value-creating technology. They then show their people how and why to use them, which is the part most organizations skip entirely.
AI also must be positioned as a tool for enabling people, not a replacement for them. Find where your teams already struggle with consistency and rework, then bring AI in to make their daily work easier. That is the airline kiosk lesson all over again. Solve the user’s problem, and acceptance follows.
FINAL THOUGHTS
One more thing worth examining: stop holding AI to a standard of perfection that we have never demanded from our existing processes. Organizations tolerate errors and inconsistency from human work every single day. But the moment someone proposes an AI tool, the first question is always, “What if it makes a mistake?” is the wrong question, and it reveals how deeply the acceptance problem runs. S = Q x A. We have been solving for the wrong variable.
George Pesansky is the president of MyBlendedLearning.com.