Ai-Measurement
- Zero Rejections and Nothing to Reject Are the Same Number
Two readers proposed different fourth failure modes for a verification gate. Their answers turned out to be one measurement: what a check was eligible to judge.
- What Is the Agent Audit? Measuring If Your AI Compounds
The Agent Audit is a two-week measurement engagement that shows, from your own git-log evidence, whether your AI gains compound or vanish by Wednesday.
- The Audit Trail: Why You Should Be Able to See Your AI's Work
Engineering-grade AI keeps an audit trail: what changed, when, and why. You trust it because you can see its work and roll it back — not on faith.
- What Is the Operator Tax in AI Workflows?
The Operator Tax is the recurring cost of re-teaching your AI the same standards every week — work that should compound but instead keeps resetting.
- You Can Measure "Judgment Work" — Here's How a Bank Would
You can measure AI output quality by scoring it against a standard you set, the way a bank scores risk. Not perfectly — usefully, enough to see a trend.
- The Month-Six Test: Open the AI Tool You Bought Last Spring
Open the AI tool you bought six months ago. Sharper than day one, or exactly the same? That one answer tells you whether you bought a system or a pile.
- The Real Enemy Isn't Your Tool — It's Drift
Your AI gets worse over time because of drift: unmeasured output that slowly decays. No new tool fixes drift on its own. Here is why, and what does.