Most conversations about AI start in the wrong place. They start with the technology: which model, which tool, which feature shipped this week. The better question, the one almost nobody asks, is the one that actually predicts outcomes. What makes one leader adopt a new technology and pull their whole organization forward, while another leader sits frozen as the world moves past them?
I spent my doctorate answering exactly that question, and the answer turns out to be remarkably stable. It does not change when the technology changes. It did not change when the technology in front of leaders was the personal computer, it did not change when it was social and collaboration platforms, and it has not changed now that it is AI. The same handful of factors decide who moves and who stalls. If you understand them, you can stop guessing about AI and start working the levers that actually matter.
The research behind this
My 2014 doctoral work was a qualitative descriptive multiple-case study of how senior leaders at Fortune 500 companies adopted social business platforms, the wave of collaboration technology that was supposed to change how large enterprises worked. It was built on in-depth interviews with 19 people: ten Fortune 500 leaders split evenly between those who had adopted and those who had resisted, plus nine employees and platform-provider staff included to triangulate the data. I analyzed all of it in NVivo to find the patterns that separated the two groups, and eight themes emerged from the interviews.
The obvious suspects did not explain who adopted. It was not budget, it was not company size, and it was not age. What separated adopters from resisters came down to a small set of factors that showed up again and again across the interviews. I have organized them here into five drivers for a business audience. The technology in that study was social platforms. Today it is AI. The drivers are the same, and most leaders get stuck on at least one.
This is not a history lesson. The pattern is playing out right now at the largest scale we have ever seen. McKinsey’s 2025 global survey found that 88 percent of organizations report using AI in at least one business function, yet only 7 percent have fully scaled it across the enterprise (McKinsey, 2025). Boston Consulting Group, surveying a thousand executives across 59 countries, found the same wall from the other side: only about a quarter of companies have moved beyond proofs of concept to generate real value, and just 4 percent are capturing substantial value (BCG, 2024). Almost everyone has the tools. Almost no one has actually adopted. That gap between having AI and using it is the entire subject of this article, and a decade ago the research already pointed at why it exists.
Driver 1: See the system, not the tools
Here is a finding from my research that surprised me, and it cuts against the obvious assumption. The adopters were not the leaders with the fewest tools. If anything they had engaged with more technology, and they offered far more examples of it than the resisters did. But engagement with tools was not what separated them. What separated them was that the adopters understood the system underneath the tools, the unified communication layer that all the individual pieces were supposed to serve. They were not collecting software. They understood the thing the software was for. The resisters, by contrast, saw a wide and growing range of options, and that dispersion left many of them with a lower understanding of what any of it actually did.
That distinction matters more than ever with AI, because the modern problem is sprawl. Every application you already use is bolting on an AI feature, and a new standalone tool launches every week with a louder promise than the last. The data on where this leads is striking: BCG found that the companies struggling with AI tend to spread themselves across six or more use cases per workflow, while the leaders pursue only about three and go deeper, rethinking the end-to-end workflow instead of bolting on tools (BCG, 2024). McKinsey reaches the same conclusion: most organizations are layering AI on top of existing processes rather than redesigning the work underneath (McKinsey, 2025). Buying more tools is not adoption. Understanding the system the tools serve is. I unpack that trap in why buying more AI tools is making you slower.
Driver 2: Understanding comes before adoption
The single cleanest predictor in my data was not money and it was not age. It was whether the leader could explain what the technology actually was. The adopting leaders could describe it in detail, give examples, and talk about how it was used. Several of the resisting leaders could not answer the basic question of what the technology even meant. One thought it referred to physical meetings. Another thought it was social media. These were not unintelligent people. They had simply never been brought up to speed, and you cannot commit to something you cannot picture.
That conclusion was direct and it has aged well. If a leader does not at least generally understand a technology, adoption does not happen, or it happens far too slowly to matter. Understanding is not a nice extra that arrives after adoption. It is the gate that has to open first. Recent work makes the same point about AI specifically: MIT Sloan Management Review argues that without leadership AI literacy and sound strategic judgment, organizations generate pilots and experimentation but never reach enterprise transformation (MIT Sloan Management Review).
That is exactly where most leaders sit with AI right now. They have heard the hype, felt the pressure, and maybe opened a chatbot once or twice. What they have not had is anyone to show them how the thing works well enough to lead with it. The pressure is high and the literacy is low, which is the worst possible combination for adoption. I go deeper on closing that specific gap in you cannot adopt what you do not understand.
Driver 3: Age is a myth, early exposure is the driver
This was the most counterintuitive finding in the entire study, and the one I repeat most often. Generation did not predict adoption. Going in, the assumption was simple and matched what nearly every participant believed: younger leaders adopt, older leaders resist. Then one participant broke the whole pattern. A Baby Boomer, squarely in the group everyone expected to resist, turned out to be one of the strongest adopters in the study. The reason had nothing to do with his age. He had loved technology since before he was ten years old and never lost the habit of staying ahead of it.
That single outlier mattered more than the averages, because it exposed what the averages were actually measuring. The real driver was early exposure and sustained openness, not birth year. Every adopting leader in the study had engaged with technology earlier in life and kept engaging. This is not just one study’s quirk. Recent peer-reviewed work across more than two thousand people found that the ability to keep up with new technology is shaped by attitudes and engagement, complex and intersecting factors that go beyond age (JMIR Aging, 2025), and Pew Research finds that about one in five adults 65 and older hold strong early-adopter preferences, the same as much younger groups (Pew Research Center).
If you have been telling yourself you are too old for AI, or that you are just not a tech person, the research retired that excuse a decade ago. Your ability to adopt does not expire with your birth year. It tracks whether you are willing to engage now. I made that case in full in you are not too old for AI.
Driver 4: Adoption starts with a recognized problem
In the study, communication was the thread that kept reappearing. The leaders who adopted tended to recognize that their communication was fragmented across too many disconnected tools, and they saw a unified platform as the fix for that specific, felt problem. Recognizing a real communication challenge and deciding to act on it was one of the conditions most associated with adoption. The leaders who could not see the fragmentation, who had no named problem driving them, had nothing to anchor adoption to.
The same holds for AI, only the pressure to do something with AI is louder than anything social platforms ever generated. So leaders buy the tools, announce the initiative, and then watch it quietly die, because it was never attached to a problem anyone actually felt. The businesses getting real value do the opposite. They start from the thing that is visibly broken, the follow-up that gets forgotten, the quote rebuilt by hand on every job, the booking that slips through because nobody had time to confirm it, and then they build the system that removes that specific pain. The technology is the last decision, not the first. I lay out that sequence in AI adoption starts with a problem, not a tool.
Driver 5: Adoption is set at the top
Leadership style was the multiplier on everything else. According to the research data, the leaders whose organizations adopted implemented clear communication, empowered their people to do the work without stressful micromanagement, and were situational in their approach, adjusting to what each moment required. The leaders whose organizations resisted tended to lead narrowly by personal example. And one conclusion from the study has stuck with me since: if the leader is unaware of the technology, the subordinates will also be unaware. Awareness did not rise up from the team. It came down from the top, or it did not come at all.
The current data says the same thing in its own language, and the numbers are stark. McKinsey found that the organizations capturing real value from AI are three times more likely than their peers to have senior leaders who own and actively champion the work (McKinsey, 2025). BCG puts an even sharper edge on it: in the companies lagging on AI, only 8 percent of C-suite leaders are deeply engaged with it, compared to nearly all of the leadership in the companies pulling ahead, and BCG names delegating AI down to middle management as a defining failure of the laggards (BCG, 2025). This is why AI is a leadership problem long before it is an IT problem. Your team’s fluency with AI is capped by yours. I make that argument directly in AI is a leadership problem, not an IT problem, and look at which leadership styles adopt fastest in situational, transactional, or transformational.
Where this leaves you
These five drivers are not a score you pass or fail. They are levers, and every one of them is yours to pull. You can learn to see the system instead of the pile of tools. You can build enough literacy to lead with confidence instead of waiting for certainty. You can drop the age excuse the research retired a decade ago. You can start from a real problem instead of a shiny tool. And you can lead the change visibly instead of delegating it and hoping. The leaders who move are not smarter or younger or better funded. They are working these levers while everyone else argues about which tool to buy.
The fastest way to see which lever is holding you back is to measure it honestly. The Omnine AI Readiness Assessment scores you across these drivers in about three minutes and points you at the single first move that will do the most. It is the most useful three minutes you can spend before you buy another tool.
References
Goodwin, M. R. (2014). A qualitative descriptive multiple-case study: Fortune 500 leaders’ social business platform adoption (Doctoral dissertation, University of Phoenix). ProQuest Dissertations Publishing (UMI No. 3648813).
McKinsey & Company. (2025). The state of AI in 2025: Agents, innovation, and transformation. McKinsey Global Survey on the state of AI.
Boston Consulting Group. (2024). Where’s the value in AI?
Boston Consulting Group. (2025). Are you generating value from AI? The widening gap.
MIT Sloan Management Review. Why AI demands a new breed of leaders.
Offerman, J., et al. (2025). Factors associated with the ability to keep up with technology developments. JMIR Aging, 8, e77930.
Pew Research Center. Older adults and technology use.