SIX
See It Work · Book 04 · AI Agents for Supply Chain · Chapter 10
Dirty data produces dirty recommendations — confidently
The defining supply-chain-agent failure: an agent built on dirty data produces dirty recommendations — confidently. The bad answer looks authoritative, so it gets acted on. The numbers are stark: the data-quality audit cost $5,000; skipping it cost $310,000. The agent worked perfectly; the data underneath it didn't.
The full detailed chart. Condensed for print legibility in the book; shown here at full size.
A confident agent on bad data is worse than no agent, because its confidence earns trust the data doesn't deserve. A cheap data-quality audit up front is the difference between a $5K check and a $310K mistake.
Operations · failure post-mortemsready
The failure is dirty data, and the audit is cheap insurance:
When Agents Fail
dirty data→ dirty recommendations
deliveredconfidently
the audit$5,000
skipping it$310,000
An agent built on dirty data produces dirty recommendations — confidently. The data-quality audit cost $5,000; skipping it cost $310,000.
For the technical reader — the command, and how to verify it yourself
# one line · you do not need to run this
see walkthrough
see walkthrough
see walkthrough
# -> the dirty-data failure and the $5K audit that prevents it
# -> the dirty-data failure and the $5K audit that prevents it
Full step-by-step is in Appendix RX: Hands-On Demonstrations in the book.