She Told Me the Training Wasn’t for Her

I had a proposal rejected last week. That happens. What stuck with me wasn’t the rejection. It was the sentence that told me it was coming before I ever heard the actual reason.

I was building a proposal for AI enablement training for the leadership and staff of an accounting firm. Partway through the call, the CEO mentioned, almost as an aside, that the training was for her team. She was too busy to sit through it herself.

That sentence told me everything I needed to know about how this engagement would go, if it went at all.

Leadership That Excludes Itself

AI is changing how accounting firms work. She knew that. She said as much on the call. But knowing it and acting on it are different things, and she was only willing to do the first one.

Any transformation broad enough to touch how a company works has to start at the top. Not because leaders need to become power users of every new tool. Because a team reads what leadership does far more closely than what leadership says.

When a CEO tells her staff this is for you, not me, she is not describing a scheduling conflict. She is telling the team how seriously to take it. The honest answer, from that point forward, is not very.

I have watched this pattern before. The initiatives that stick are the ones where a leader shows up first, asks the harder questions in the room, and admits what she does not yet understand. The initiatives that stall are the ones where leadership signs the check and steps out of the room.

It Was Never Just the Team’s Problem

Here is the part that made this case harder to walk away from politely. AI does not usually eliminate jobs. It takes over tasks.

That distinction matters more than it sounds like it should. When a piece of work gets automated, the job built around that work does not disappear. It changes. Responsibilities shift. Titles stop matching what people actually do day to day. Some roles expand. Some narrow. All of that needs a decision maker who can see the whole org chart, not just the tasks in front of one department.

A CEO is exactly that person. Which means the moment she said this technology would touch every job except hers, she had already misjudged what the technology does. It does not carve out an exception for the corner office. If anything, the corner office is where the most consequential decisions about role redesign get made.

I was not just proposing a training anymore. I was going to have to make the case for why the person paying for the room needed to be in it.

We Already Have an IT Company for That

That was not, in the end, why the proposal died. It died because she told me the firm already had a company managing their IT infrastructure, and that covered it.

This is a mistake I see often enough that it deserves its own name. Treating an AI initiative like an IT purchase.

An IT vendor manages infrastructure. Servers, networks, software licenses, security patches. That is real work, and it matters. But it is not the same work as deciding how tasks, roles, and responsibilities should change across a business.

An AI initiative needs an executive sponsor for the same reason a merger or a new service line needs one. It reaches into how people spend their time and what their jobs are for. That is a business decision. It has to be owned at the business unit level or higher, by someone with the authority to redesign how the company operates, not just the authority to keep the network running.

Handing that decision to an IT manager is not a shortcut. It is a mismatch between the size of the decision and the level of the person making it.

One Mistake, Told Twice

Both reasons this deal fell apart were the same mistake in different clothes. First, the CEO handed the training to her staff and stepped back. Then she handed the whole initiative to a vendor who was never positioned to own it.

Neither decision was really about AI. Both were about who gets to decide what changes, and who gets to stay out of the room.

If your AI initiative does not need you in the room, it is probably not the kind of initiative that will change anything.