An anonymous real example
An HR team used AI to interpret employee engagement feedback. The summary looked plausible, but experienced colleagues felt the sentiment and nuances were not quite right. Their knowledge of the context made them stop and check. That opened a more difficult question: would someone with less experience have recognised the same problem? The team could see that the value of the output depended partly on judgement that did not appear anywhere in the output itself.
The assumption worth revisiting.
If experienced people make AI-assisted work look easy, it is tempting to assume the experience is no longer needed. In reality, it may be doing more of the quality control than the process recognises. The assumption to unlearn is that less time producing an answer means less expertise is needed to own it. Production and judgement are different capabilities. You also need a way for the next generation to develop that judgement. A misleading interpretation of employee feedback could send leaders towards the wrong intervention. More widely, removing experienced reviewers before others can recognise errors can weaken decisions while the volume of output rises.
Which expertise is making this look easier than it really is?
Give the capacity a purpose.
Protect time for experienced colleagues to explain what they noticed, how they checked it and what changed. Use released capacity to build judgement in others before assuming the organisation can operate with less of it.
How this fits the capacity allocation →
Use AI to challenge your first interpretation
Explore this question without assuming our first interpretation is right: Which expertise is making this look easier than it really is? Offer competing explanations, identify assumptions and suggest evidence we could obtain from the people doing the work.
Treat the answers as possibilities to test with people and evidence.
Go deeper into the evidence.
Anthropic Research · Research explainer and linked randomised-study preprint
A small experiment recruited 52 mostly junior Python engineers learning an unfamiliar library. The AI group averaged 50% on an immediate comprehension quiz versus 67% without AI: a 17-percentage-point gap. The approximately two-minute completion-time difference was not statistically significant. Explanatory and conceptual uses of AI were associated with stronger understanding than wholesale delegation.
Short task and immediate quiz, not a longitudinal workplace study. Interaction-pattern comparisons are observational. The source's shorthand '17% lower' should be rendered as 17 percentage points using its reported means.
ACM CHI 2025; Microsoft Research · Peer-reviewed survey
319 knowledge workers supplied 936 examples of AI use. Greater confidence in AI was associated with less reported critical-thinking effort; confidence in one's own ability was associated with more. Thinking shifted toward checking information, integrating output and overseeing the task.
Self-reported correlations. This does not prove AI causes declining intelligence or long-term skill loss.
podcast · 33 min 27 sec
From AI experimentation to AI maturity
CIPD
Connects adoption with workforce readiness and returns. Useful when leaders have bought tools but cannot yet identify dependable capacity.
Listen or watch for: Which missing capability is stopping experimentation becoming a repeatable way of working?
About this resource
Practitioner discussion, including a Nationwide representative; it is not an independent evaluation of productivity gains.