The People Side of AI Change

In most of the rollouts I've worked on, the implementation plan finishes at go-live. The gap that opens up next — between the tool being live and your team actually using it every day — is called the adoption gap.

What's sitting in that gap is a people and change management question, not a technology one. The ROI your organisation is expecting only arrives once your team is using the new system, process or tool the way it was designed to be used.

Projects are far more likely to meet their objectives when people are supported to understand, adopt and use the change. Prosci's benchmarking research found that 88% of projects with excellent change management met or exceeded their objectives, compared with 13% where change management was poor — a difference of roughly six to seven times.

Here's what strong adoption looks like in the day-to-day work that follows go-live:

Your rollout gains momentum when you stay connected to your team after go-live — paying attention to their experience of the change.

Indeed's 2026 AU Jobs & Hiring Trends Report found the same pattern nationally: 32% of Australian workers now use AI multiple times a week, and an equal 32% have disengaged from it entirely. That leaves 36% of the workforce still deciding — not locked into either camp, and still open to influence. The Prosci figures above and this split are telling the same story: change management, not the tool itself, decides who ends up using AI well. This course teaches that methodology, mapped out in the framework below — bringing both the 36% still deciding and the 32% who've disengaged into real, daily adoption.

Many AI projects begin with a capable tool, a committed vendor and a clear business case. The results depend on what happens next: how well the organisation helps people understand the change, see its relevance and build it into their work.

MIT research published in 2025 found that 95% of enterprise generative AI pilots delivered no measurable impact on the bottom line. The opportunity sits in how organisations integrate the tools into real work, not simply in making the tools available.

This is where communication, training and day-to-day leadership matter. David Rock's SCARF model describes five social needs people continually look for at work:

• Status — am I still valued here?
• Certainty — can I see what is coming?
• Autonomy — do I have a say in this?
• Relatedness — do I still belong in this team?
• Fairness — is this being handled fairly?

An AI announcement can raise questions across several of these areas before anyone has opened a demo link.
A change to someone's role or way of working may feel highly personal, and that sense of threat can narrow attention and make it harder to take in new information.

The practical leadership task is to create enough clarity, involvement and support for people to stay open to learning — this is good change management.
When you reduce the social threat around a change, you create more capacity for people to engage, practise and adapt.

"We've done technology change before — ERP, cloud, remote working tools. We know how this goes."

Usually that would be true. This time the territory is different.

Previous technology change replaced administrative and manual tasks — things people were largely glad to hand off. AI moves into different territory, because it can potentially take on:

• writing first drafts, editing, and refining content
• researching topics and synthesising findings
• analysing data and identifying patterns
• generating advice, options, and recommendations
• building presentations, proposals, and reports
• preparing for meetings and client conversations
• creating ideas and solving problems
• designing training content and learning materials
• reviewing and summarising documents

These are not peripheral tasks. For many people these are the job itself — skills they spent years building, and the thing that defines them professionally and justifies their place in the team.

When AI can do the drafting, what does it mean to be the drafter? What does it mean to be the analyst when AI can run the analysis? That question surfaces in your team whether you name it or not — as silence, resistance, or performative use where people appear to engage while changing nothing.

Here's what the research actually shows: AI is not replacing people — it's changing what people are asked to do. The World Economic Forum's Future of Jobs research projects that AI will create more roles than it displaces. What is happening is a shift in what expertise looks like — and that shift needs managing, not ignoring.

Here's a rollout pattern you've probably seen — or run:

There's a name for this: the tool reveal. It gets the announcement done, and it skips every condition that actually makes adoption happen.

A demo shows people what's possible. It doesn't give them:

1. A reason tied to their own workload
2. Time to practise on real tasks
3. Permission to be clumsy with it at first

Without those three things, watching a tool work is not the same as using it.

We’ll get to those conditions properly in the framework ahead — first, what do you think the core assumption was?

Everything in this course maps to the 5-step change methodology I've built across 20 years of change practice and more than 50 technology projects. It's the framework behind Change Made Simple's free AI tools and consulting, and it runs in sequence — each step creates the conditions for the next. Click each step to see what it involves.