I led our agency through its restructuring during our AI transition, so I'll say plainly what the vendors won't: AI transformation is mostly a leadership challenge, not a technology one. The models are the easy part. The hard part is people, incentives, fear, and decision-making, and none of that lives in the IT department.

When a transformation stalls, leaders tend to blame the tools or the team. In my experience the real bottleneck is almost always at the top: unclear priorities, mixed signals about job security, and a refusal to change how the organization actually works.

Why the conventional wisdom is wrong

The conventional approach buys a platform, assigns it to the tech team, and waits for transformation to happen. It treats AI as software to install. But AI changes how people work, what their roles mean, and who holds power, and you cannot delegate that kind of change to a tools team. It belongs to leadership.

What is actually true

AI transformation succeeds when leadership does the human work: setting a clear direction, being honest about how roles will change, redesigning processes and incentives, and modeling adoption from the top. The technology follows. Skip the leadership work and the best platform in the world will sit unused while people quietly protect the old way.

People don't resist tools. They resist uncertainty, lost status, and change imposed without explanation. That's a leadership problem with a leadership solution. When a capable team won't adopt a perfectly good system, it is almost never because the system is hard. It is because no one has told them honestly what it means for them, so they assume the worst and dig in.

There's also an incentive problem leaders forget to look at. If you ask people to embrace AI while their bonuses, targets, and status still reward the old manual way of working, they will rationally keep doing the old thing. Transformation requires rewiring what the organization actually rewards, and only leadership can touch those levers.

What we learned at TTGC

Our own transition taught me this firsthand. The technical rollout was straightforward; the organizational change was not. People needed to know their roles were evolving, not vanishing, and they needed to see leadership using the same tools and changing the same habits we asked of them. Restructuring the org, naming new roles, and being honest about what was changing mattered far more than any model we adopted. When we now guide clients, we spend more time in the leadership room than the server room, because that is where transformations are won or lost.

The hardest and most important conversations were never about technology. They were about people: who does what now, what happens to roles the AI touches, and how we keep trust while everything shifts. Leaders who avoided those conversations to seem decisive on tooling found their rollouts quietly stalling. Leaders who had them, even when they were uncomfortable, got teams that moved with them instead of against them.

The honest take

If your AI transformation is stalling, don't look at the technology first. Look at whether leadership has set a clear direction, addressed people's fears, and changed how the organization actually operates. Buying AI is easy. Leading people through what it changes is the real work, and it cannot be outsourced to a tools team. Own it from the top, or watch it stall.

Sources