I recently had the privilege of leading a series of executive sessions on GenAI at the University of Oxford. The format was intimate by design — small roundtables of six to eight C-level leaders at a time, less lecture and more working conversation.
Walking in, I expected questions about models, platforms, and capabilities. These were sophisticated leaders, many already running enterprise AI tools across tens of thousands of employees. They didn't need a primer.
What struck me was how consistently the conversation landed on exactly four things — and none of them were technology.
Leadership. The organizations making real progress aren't the ones where executives sponsor AI initiatives from a distance. They're the ones where leaders are catalysts — using the tools themselves, talking about what worked and what didn't, and modeling the curiosity they want their teams to have. Sponsorship delegates. Catalysis demonstrates.
Enablement. Access is not adoption. Nearly every leader in the room had already rolled out AI tools broadly. The harder question was what happens after the license is provisioned. The gap between "everyone has it" and "everyone gets value from it" is closed by enablement: role-specific training, time and permission to experiment, and clear examples of what good looks like in each function.
Champion networks. Change doesn't scale through mandates; it scales through people. The pattern that came up again and again was the power of identifying the early enthusiasts — the ones already finding clever uses on their own — and connecting them into a deliberate network. Champions translate the abstract into the local: "here's how this works in our team, on our problems." No central program can replicate that credibility.
Organizational governance. This one surprised me most — not that it came up, but how it came up. These leaders weren't treating governance as a brake. They saw it as the thing that makes speed possible: clear guardrails on data, clear decision rights on use cases, clear accountability for outcomes. Teams move faster when they know where the edges are. Ambiguity, not oversight, is what slows adoption down.
What ties these four together is a simple idea: the constraint on AI value is no longer the technology. The models are capable. The tools are available. The constraint is organizational — how people are led, equipped, connected, and governed.
That's actually good news. It means the biggest levers are ones every leadership team already controls. You don't need to wait for the next model release to make progress. You need leaders who go first, enablement that meets people where they work, champions who carry the change peer to peer, and governance that gives everyone confidence to move.
AI presents enormous opportunity. But the opportunity isn't captured in the procurement decision or the pilot. It's captured in the thousands of daily decisions people make about whether and how to use it — and those decisions are shaped by exactly the four things these executives kept coming back to.
The technology is ready. The question every leadership team should be asking is: are we?
