Core insight: organizations accrue 'learning debt'—unverified assumptions and aging beliefs—that quietly bias strategy, hiring, and automation. Left unmanaged, these stale models compound into expensive bets. The Learning Debt Ledger is a minimal habit you run in under ten minutes: capture the critical assumption, the smallest falsifying test, an owner with a short deadline, and a decay rule that forces re‑test or retirement. In this episode I give the exact one‑line Ledger entry, three enforcement patterns (test windows, evidence gates for commitments, and public debt rollups), and two constrained AI prompts to 1) convert meeting notes into candidate ledger rows and 2) draft a 7‑day micro‑experiment that will falsify or reprioritize the assumption. You get a 7‑day pilot to replace guessing with measurable tests and one immediate action: add three high‑risk assumptions to your Ledger before the end of the week. CTA: subscribe. Stay agentic.