The Quiet Risk No One Is Planning For
What happens if organisations stop producing their future experts?
Flagship Essay — Capability Architecture Series
The productivity case for AI copilots is not hard to see. The appeal is easy to understand. The briefing note appears faster. The contract summary takes minutes. The first draft is ready before the meeting has finished. A senior person can ask for the comparison, the synthesis and the tidy next-step email without waiting for someone junior to pull it together by hand.
That is useful.
But there is a quieter question underneath the efficiency gain.
What if some of the work organisations are most excited to remove is the work people used to learn from?
Not the glamorous work. Not the work anyone put on a career brochure.
The messy first pass. The weak draft. The boring comparison table. The document review where someone slowly learned what mattered and what was noise.
AI is starting to take more of that work.
It may be the moment organisations accidentally make the future expert pipeline thinner while everyone is congratulating themselves on throughput.
Always a fun little governance surprise.
The Shift
The specific AI capability here is not vague “automation.”
It is embedded workflow AI: copilots and assistants that draft, summarise, compare, analyse, review documents, prepare notes and produce first-pass thinking inside the tools people already use.
AI is no longer only helping people do the work. In some workflows, it is doing the first pass, the messy pass and sometimes most of the pass.
That matters because first-pass work has never only been output. It has also been apprenticeship.
People learn by doing the clumsy version first. They learn what a bad brief looks like by writing one. They learn what a weak argument feels like by trying to defend it. They learn which details matter by including too many and having someone more experienced quietly circle the three that count.
Not always gently, admittedly. Many of us have the tracked changes to prove it.
But that repetition did something.
It built judgement.
The Real Tension
Most organisations are not trying to weaken their future capability.
They are trying to move faster, reduce cost, improve consistency and make better use of experienced people. That logic is reasonable.
The tension is that the short-term productivity gain is visible, while the long-term capability loss is harder to see.
If AI absorbs routine drafting, research, synthesis and document review, the organisation may still get the work done. It may even get it done better.
But who is learning how the work works?
A junior employee who only reviews AI output may move faster. They may also skip the part where pattern recognition forms. They can approve a summary without wrestling with the raw material. They can edit a recommendation without knowing how the options were weighed.
That is the supervision paradox.
We may ask people to check work they have never really done.
And then, a few years later, wonder why the mid-career bench feels thinner than expected.
The Ripple Insight
This is not mainly a jobs argument.
It is a question about how organisations keep building judgement.
Entry-level roles have always done two jobs. They produce work today, and they produce the people who will carry harder work tomorrow.
The first job is easy to measure. The second is easier to forget.
That is why the AI productivity conversation can become misleading. It asks how much faster the task can be completed. Fair question. But it often misses the better one:
what developmental work was hidden inside that task?
Because sometimes the inefficiency was the point.
The slow draft, the repeated review, the awkward first client note, the analysis rebuilt after a senior person asked one annoying but correct question — these were not just delays in the system. They were how the system produced judgement.
No one wants to preserve busywork for sentimental reasons. I certainly do not.
But leaders need to know the difference between removing waste and removing practice.
Those are not the same thing.
If organisations remove repetitive low-value work, good. Please do. There is plenty of it.
But if they remove the early exposure that teaches people how to think, interpret, challenge and recover, they may be borrowing capability from the future.
And the future, irritatingly, tends to send the invoice later.
The Move
Before automating or compressing early-career work, leaders should ask a more concrete question:
Where does judgement currently form?
Not in the values statement. Not in the learning platform. In the actual work.
Pick one function, role or pathway. Look at the tasks AI is now absorbing: drafting, analysis, research, document review, meeting preparation, customer summaries, case notes, code review, contract triage.
Then separate the work into two piles.
One pile is genuine waste: repetitive, low-learning work that nobody needs to defend for sentimental reasons.
The other pile is different. It may look inefficient, but it teaches pattern recognition, context, trade-offs and professional instinct.
That second pile needs redesign, not disappearance.
It may mean structured first-pass ownership before AI assistance. It may mean requiring juniors to explain what the AI missed. It may mean pairing AI output with senior review conversations or rotating people through messy real cases before they supervise polished machine output.
The goal is not to slow the organisation down for the romance of apprenticeship.
The goal is to keep producing people who can make good calls when the answer is not in the tool.
The quiet risk is not that AI makes people faster.
It is that organisations may stop noticing where expertise used to be made.
Executive note for leaders
Before removing a task with AI, ask:
is this task only producing output, or is it also producing judgement?
If it is only output, automate it carefully.
If it is also building judgement, pause before removing it and redesign the pathway around it.
Identify where first drafts, analysis and review are disappearing — then decide how future experts will get the repetition and feedback they still need.
Organisations do not inherit expertise.
They design the conditions under which it forms.
Subscribe
You can subscribe for free. Every edition is delivered directly to your inbox and published here on Substack.
If a piece raises a question, surfaces a pattern, or helps you think more clearly about a decision, I’d value the conversation.
Thanks for reading,
Stuart Gonsal MAICD
With occasional help from Springsteen, my Border Collie, who reminds me that clarity comes from movement 🐾.
Connect
LinkedIn – Follow for practical leadership in the AI era. ↗
Disclaimer
Everything shared in The Ripple Effect reflects my personal views and does not reflect those of my current or past employers, clients or partners. Any examples are illustrative, drawn from publicly known patterns or anonymised experience.