Frederick Winslow Taylor's scientific management (1911) treated humans as machine components — measure everything, eliminate waste, maximise throughput. It was not wrong about efficiency. It was wrong about people.
Decades of HRM research established load-bearing truths: intrinsic motivation outperforms extrinsic pressure over time; recovery time is not wasted time; autonomy, mastery, and purpose are not soft perks — they are conditions for sustained performance.
AI is being sold as 'now you can do more,' not 'do the same, better'. The implicit contract is expansion. You get the tool; you give back the time savings in the form of higher output. The ceiling rises. So do the expectations.
"Dead time" stops being a category. Every moment becomes potentially productive. The grocery run is a voice note. The commute is a review cycle. The capability expands to fill the available space — without anyone planning it that way.
You can delegate execution. You cannot delegate judgment. Accountability for quality, timing, and whether the thing made sense in the first place — none of that transfers to the agent. The speed increases. The cognitive load of oversight does not decrease at the same rate.
The Industrial Revolution eventually produced the 8-hour day and the weekend — not because capitalism decided to be kind, but because burnout and instability produced outcomes even efficiency-maximisers found unacceptable. The question is how long the AI equivalent takes.
"We spent a century learning that the goal isn't maximum output per worker-hour. The goal is sustainable, meaningful, high-quality work. And yet."
Three forces that compound when AI capability expands unchecked. Recognising the loop is the first step to breaking it.