The Candy Machine Trap
What it is
The Candy Machine Trap is the named pattern of a leader’s AI hours flowing overwhelmingly into the Candy Machine quadrant of the AI Allocation Matrix (companion glossary entry): shallow, maintenance work like drafting routine emails, polishing slides, summarising meetings. The compounding-leverage quadrant (strategic framing, decision diagnosis, judgement-heavy writing, developmental conversation) sits empty. The candy-machine analogy captures the appeal: cheap, quick, immediate, easily repeated. Two failure modes flow from the Trap: lopsided allocation, and invisible displacement of the hour that might have compounded.
What it is
The Candy Machine Trap is the named pattern of a leader’s AI hours flowing overwhelmingly into the Candy Machine quadrant of the AI Allocation Matrix (companion glossary entry): shallow, maintenance work like drafting routine emails, polishing slides, summarising meetings. The compounding-leverage quadrant (strategic framing, decision diagnosis, judgement-heavy writing, developmental conversation) sits empty. The candy-machine analogy captures the appeal: cheap, quick, immediate, easily repeated. Two failure modes flow from the Trap: lopsided allocation, and invisible displacement of the hour that might have compounded.
Why it happens with AI
AI removes the activation energy for shallow-maintenance tasks and leaves the activation energy for deep-investment work unchanged. The leader still has to think, frame and judge for the strategic question; the routine email writes itself. The result is structural. AI’s frictionlessness pulls hours from where leverage lives toward where it does not. Sir Thomas Gresham’s law of money (1558), bad money drives out good, applied to work names the mechanism: easy work drives out hard work whenever the leader has discretion over time, accelerated by a frictionless tool.
What working on it does, impact and benefits
Naming the trap and tracking where the hours go converts an invisible displacement into a visible one. Leaders who do this reclaim the compounding-leverage hour the Candy Machine had been quietly absorbing: the strategic diagnosis, the judgement-heavy memo, the developmental conversation. Compounding returns concentrate in those hours; they do not show up in the Candy-Machine ones, no matter how many AI prompts get logged. The benefit is not less AI use; it is AI use that produces the thinking the role demands.
Canonical framework: virenlall.com/candy-machine-trap, the full ~600-word treatment of the Trap, the AI Allocation Matrix, the three corrective moves and the three embedded habits.
“AI gravitates to the work that matters least, away from the work that matters most.” — Viren Lall, Managing Director, ChangeSchool LDN (2026).