AI Won't Replace Change Managers. The Role Is Changing.

We have been publishing a lot of articles in recent weeks and as I sat at our weekly CCG Advisory meeting yesterday, I wondered what I would be able to cook up this week. As I pondered this and became increasingly despondent, a thought immediately became apparent to me. The burning question that all consulting firms are considering at the moment: “Will AI replace all of us?” “Will AI replace all the hours of decks and planning and programming and client meetings that try to do change better?”

At CCG we have spent more than twenty years arguing that “change management” is the wrong name for what we do. The plan was never the point. Nor was the training, or the deck. The real work was always helping people negotiate the messy, political, deeply human business of doing something new.

I think I have the answer to these two very difficult questions and I think that AI has just proved our point for us.

The machine has taken the mechanical half

Here is what is actually happening in our profession. The tools now write the first communication. They summarise the survey. They build the deck in the time it takes to make a coffee. Ask whether a machine can “do change management” and the honest answer, this year, is: more of it every month.

So we would ask the question we have always asked instead. Which part of this work was ever the point?

Not the deck. We have never believed the deck was the point.

Prosci went and looked at what change practitioners actually use AI for.¹ The answers fall into five tidy piles: communications, content, planning, automation, analysis. Read that list slowly. Not one of those piles is trust. Not one is politics. Not one is the quiet conversation with a sponsor who has stopped believing and has not yet told anyone.

That is not a gap in the technology. That is the whole of our discipline, sitting untouched.

(There is a lovely detail buried in the same research. Familiarity with AI among practitioners has actually gone backwards over the year, from 84% to 77%.¹ Eighteen months of noise, and our profession is quietly getting on with the part of the job that was never going to be automated anyway.)

Think of the surgeon

Here is how we explain it to clients. A surgeon can now hand the charting, the scheduling, a good deal of the diagnosis to a machine. Nobody imagines this makes the surgeon less necessary. It makes the judgement, when to cut and when to wait, when to tell a family the truth, the entire job.

Change work is no different. AI has taken the stitching. It has not taken the decision.

And why would anyone argue with the surgeon about the surgery, yet cheerfully argue with a change consultant about the change? We have never quite worked that out. But it is exactly the same instinct, and it is exactly as expensive.

The mistake we keep watching organisations make

And yet. Look around and you will see companies doing something strange. They are using AI as the reason to let go of the very people who hold that judgement.

Gartner asked 350 large organisations already running these systems, and found that roughly four in five had cut staff to pay for them.² It did not work. The ones who cut were no more likely to see a return than the ones who kept their nerve.

We are not remotely surprised. We have watched this film before. Strip out the people who understand the complexity, at the very moment the complexity increases, and you do not save money. You simply lose the only people who could have made the thing work.

What is left is the part we always cared about

So what remains for the change professional, once the machine has taken the mechanical half? The hard half. The half we have always insisted was the real one.

Reading a room. Holding an uncomfortable truth in front of a nervous executive. Deciding, when a system flags twelve people as “resistant”, whether that is a redeployment, a coaching conversation, or simply a bad week that needs nothing at all.

A model has no reputation to lose and no team to face on a Monday morning. When a change lands badly, a person carries it. That is not a flaw in the human role. In our experience, it is the human role.

None of this is new to us. It is the argument we have made since we first started calling our work strategy implementation rather than change management. What is new is that the tools have finally drawn the line for everyone to see. On one side, the mechanical work, now largely done by a machine. On the other, the judgement, the trust, the courage to have the difficult conversation. That second side is not shrinking. It is the whole game.

A gentle challenge to anyone leading change

So we would gently suggest you stop asking whether AI will replace your change function. Ask instead which of its tasks the machine has already quietly taken, whether anyone decided that or not. Then ask the harder question. Have you put the time it freed back into judgement, or simply pocketed it?

Most organisations have never thought to ask. The ones who will do well are the ones prepared to treat their own change team the way we ask them to treat the whole business. As a system to be redesigned on purpose, not left to drift.

This is the work we do, and, in truth, the work we have always done. CCG Learn builds the judgement this moment demands. CCG Consult redesigns the change function so that roles and decision rights match what the machine now handles. CCG Analytics measures whether people have genuinely come with you, rather than assuming that use means acceptance. And CCG Advisory helps boards answer the question sitting underneath all of it: who is accountable when a system, and not a person, produced the plan?

I can breathe more easily today and answer the question as follows.

Will AI replace change managers? No. But it is quietly clearing away the parts of the job that were never really the point, and handing us back the part that always was. We think that is rather good news.

If you would like help with Doing Change Better, we would welcome the conversation. Get in touch with us.