Lean made factories reliable. The same discipline can make AI reliable.
Taiichi Ohno catalogued the seven Muda: transportation, inventory, motion, waiting, overproduction, overprocessing, and defects. Each was a category of activity that consumed resources without adding value. Shigeo Shingo added poka-yoke, error-proofing: a step designed so the defect couldn't form in the first place. The world changed when enough plants learned to see and remove these systematically.
The lesson wasn't manufacturing. It was that quality is the shape of the process, not a checkpoint at the end of it.
Quality is not a checkpoint. It is the shape of the process.
Proxies don't move pallets, but a fleet of them generates its own distinctive waste. Permission Waste: cases blocked because nobody said yes. Capability Waste: idle apprentices beside a growing backlog. Exposure Waste: authority left hot after the scenario closed. Transport Waste: handoffs that add no reasoning. Defect Waste: decisions reopened because the data, the reasoning, or the coordination produced an outcome no one would have chosen.
Each one becomes visible the moment the fleet is instrumented for it. Reading those signals is how stewards decide where the next sigma of improvement comes from.
A veto lens that prevents a decision from progressing when a required signal is missing is poka-yoke. An authority cap that refuses to escalate without a scenario change is poka-yoke. A premise-staleness check that pauses a case when its underlying conditions have shifted is poka-yoke. None of these are audit steps after the fact. They are structure that forbids the wrong outcome from forming in the first place.
This is the discipline Lean asked of manufacturing, and that the book asks of AI.
Including Chapter 22 (A Six Sigma for Judgment) and Appendix D (the five categories of kinetic waste). By Christopher Jackson, May 2026.
Read about the book arrow_forwardNo spam. One email when Proxy.Me is available.