Comfort, learning, panic

There is an old model from adult learning that I keep returning to, because it explains more about enterprise AI adoption than any technology framework I know¹. It says people operate in one of three zones. The comfort zone, where everything is familiar and nothing is learned. The learning zone, where there is enough challenge to grow and enough safety to try. And the panic zone, where the challenge overwhelms the safety, and learning shuts down entirely — replaced by defence, avoidance and performance theatre.

Most organisational AI strategies, judged by their effects rather than their intentions, are machines for catapulting people from the comfort zone directly into the panic zone, clean over the learning zone in the middle.

Consider the standard sequence. For months, nothing: no tools, no guidance, a vague policy discouraging experimentation. Comfort zone, enforced. Then a strategy lands — an announcement with the words "AI-first", a rollout, a town hall where someone says roles will change, an unspoken message that everyone should already be across this. Panic zone, at scale. The middle bit — the structured, supported, deliberately safe environment where people could actually build capability — was never built. It was assumed, or worse, outsourced to a licence.

Then leadership is puzzled by the response: the quiet refusal to adopt, the senior people who delegate all AI matters downward, the compliance-flavoured enthusiasm that produces activity and no change. None of this is irrational. It is what humans do in the panic zone. The behaviour is diagnostic, and what it diagnoses is the missing middle.

Why AI hits the panic threshold faster

It is worth being precise about why this particular technology trips the panic response in ways the last several did not.

Cloud, mobile, even the internet changed what organisations could do, but they mostly did not challenge what senior professionals were. AI does. When a model produces in forty seconds a version of the analysis you built a career on, the threat is not to your toolkit — it is to your identity. Twenty years of accumulated expertise is the asset your status, income and self-understanding rest on, and it is being visibly repriced in real time. That is not a training gap. That is an existential event, arriving at work, unannounced.

And the people most exposed are precisely the most senior. A graduate has little expertise to defend and everything to gain. An executive has decades of it, plus an audience. Which produces the pattern I see constantly in coaching: the more senior the leader, the less safe it is to be a beginner. There is nowhere to practise being bad at the new thing. Every meeting is a stage. Asking a basic question about how these systems work feels like an admission with career consequences. So the most influential people in the organisation — the ones whose modelling of learning behaviour would matter most — quietly opt out, and their opting out cascades.

The panic zone has an organisational shape too, not just an individual one. Announcing transformation while providing no protected practice time; measuring adoption weekly while offering training annually; celebrating AI wins in the same quarter as an AI-attributed restructure — each of these raises threat and lowers safety. The learning zone is not a mood. It is a set of concrete conditions, and they are either provided or they are not.

What the learning zone actually requires

The conditions are knowable. Decades of research on adult skill acquisition, and the last three years of watching organisations do this well and badly, point the same way.

Real work, low stakes. People do not build AI judgement on toy exercises, and they cannot build it on live deliverables where an error is career-limiting. The learning zone is real workflows with the stakes lowered: sandboxed versions of actual tasks, paired working, outputs that get reviewed before they ship. Two days spent working your own material with an expert in the room outperforms twenty hours of e-learning by an order of magnitude, and it is not close.

Permission that is visible, specific and senior. A policy that permits experimentation is worth little. A divisional head who opens a session with the prompt they wrote badly last week, what went wrong, and what they changed — that is worth a great deal. Psychological safety around AI is created almost entirely by senior people performing beginnerhood in public. It is the cheapest intervention available and the least used.

Time that is actually protected. The learning zone costs hours, and the hours are the first thing operational pressure reclaims. If AI capability-building is not scheduled, resourced and defended like any other critical program, the organisation has chosen the comfort zone while announcing the strategy. The early weeks of AI-assisted work are often slower than the old way — the fluency dip is real — and organisations that do not budget for the dip conclude the technology failed, when what failed was the patience.

Honesty about the stakes. This is the counterintuitive one. It is tempting to soothe: nobody's job will change, this is just a tool. People do not believe it, and the gap between the reassurance and what they can see with their own eyes increases threat. What lowers threat is honesty with agency attached: some of how we work will change substantially; here is what we can see, here is what we cannot, and here is the support that will be there as it changes. Adults calibrate to honest uncertainty far better than to incredible comfort.

The leader's own zone

There is a version of this essay that is about your organisation. But the version that matters more, if you are a senior leader, is about you.

Where are you, personally? Comfort — deferring the topic, delegating it, sponsoring it from a distance? Panic — privately convinced you are too far behind to start, publicly fluent in the vocabulary? Or genuinely learning — hands on the tools, regularly, on your own work, with someone who can tell you the truth about your blind spots?

The question is not rhetorical, because your zone sets the available zones for everyone below you. A leadership team in comfort licenses comfort. A leadership team in panic transmits panic — organisations are exquisite amplifiers of executive anxiety. A leadership team visibly in the learning zone — practising, erring, improving, talking about it — makes learning the culturally safe default. Of all the leverage points in an enterprise AI transition, the leadership team's own learning posture is the one with the highest ratio of impact to cost, and it is the one most often skipped, because it is the one that cannot be delegated.

The strategic argument for all this is simple. The organisations that win the next decade will not be the ones that adopted tools fastest. They will be the ones that learned fastest — because the technology will keep moving, and the durable advantage is not any particular capability but the organisational muscle of acquiring the next one. That muscle is built in exactly one place: the zone between comfort and panic. Building it, deliberately, for real executives with real stakes, is most of what our practice does. If your organisation is stuck in one of the outer zones, we should talk.

References

Footnotes

1. ThemPra Social Pedagogy. (n.d.). The learning zone model. https://www.thempra.org.uk/social-pedagogy/key-concepts-in-social-pedagogy/the-learning-zone-model/ ↩