When Adoption Creates the Illusion of Progress
At Inter IKEA, around 80 percent of communicators were already using AI regularly.
In most organisations, that level of adoption would be seen as a milestone.
Instead, it raised a question.
If nearly everyone is using AI, why does it still feel like nothing has fundamentally changed?
The answer lies in how AI is being used.
Most of the activity sits at the surface. Drafting emails. Summarising documents. Translating content. Fixing tone. These are useful improvements, but they do not reshape the system. They make existing processes faster without challenging how those processes are designed.
It is progress, but not transformation.
The Cost of Letting AI Grow Organically
Left unchecked, this kind of adoption creates fragmentation.
Different teams experiment in different ways. Some become highly efficient, others lag behind. Outputs vary. The same tasks are repeated across the organisation because no one knows what has already been tested elsewhere.
Over time, a pattern emerges.
AI becomes dependent on a handful of enthusiastic individuals. Knowledge remains informal. Questions about what is acceptable and what is not continue to resurface. And perhaps most importantly, there is no clear way to measure what any of it is actually delivering.
What started as innovation begins to look like noise.
The Moment AI Stops Being a Tool
The turning point comes when organisations realise that AI is not just a tool.
It is an operating model question.
Using AI well is not about writing better prompts or finding the right platform. It is about redefining how work flows through the organisation. How content is created, reviewed and approved. How decisions are informed. How teams collaborate.
This is where most organisations hesitate.
Because changing tools is easy. Changing how a function operates is not.
Why Scaling AI Requires Restraint
One of the more counterintuitive lessons from IKEA’s approach is that scaling AI is not about moving faster.
It is about moving deliberately.
Instead of trying to implement AI everywhere at once, the focus was placed on building depth in specific areas. At the same time, the transformation was driven from the bottom up, grounded in real use cases rather than imposed as a top down mandate.
This creates something that most AI strategies lack.
Coherence.
Without it, organisations move quickly but in different directions. With it, progress compounds.
The Human Factor No Strategy Can Ignore
Technology tends to dominate the conversation around AI.
But the real constraint is rarely technological.
It is human.
There is an underlying tension inside most organisations. AI promises efficiency, but it also raises uncomfortable questions. About roles. About value. About what happens next.
If those questions are not addressed directly, they slow everything down.
That is why the role of leadership becomes critical. Not just in setting direction, but in shaping mindset. Creating clarity. Building confidence. Making it clear that AI is not replacing communicators, but redefining what great communication looks like.
Without that shift, adoption remains cautious and fragmented.
The Work Is Only Beginning
Even for organisations that are ahead, the journey is far from complete.
Pilots need to be scaled. Processes need to be refined. New ways of working need to be tested and adjusted continuously.
There is no finish line.
Only iteration.
What This Means for Communications Leaders
The biggest risk today is not falling behind on AI.
It is believing that adoption equals transformation.
Most teams are already using AI. That is no longer the differentiator.
The real question is whether AI is changing how the organisation operates or simply making existing habits more efficient.
Because in the next phase, the advantage will not belong to the teams that use AI the most.
It will belong to the ones that use it with intent.
And that is a very different challenge.
Our next AI in Communications Boot Camp takes place from 4 to 5 June in Calgary, where we will continue exploring how communications teams can move from experimentation to fully integrated AI operating models.