---
id: "question-maintaining-codified-judgment"
type: "open-question"
source_timestamps: ["§ What Separates Leaders from Laggards"]
tags: ["maintenance", "lifecycle-management"]
related: ["concept-codifying-judgment"]
resolutionPath: "Develop a lifecycle management framework for 'judgment files,' including regular audit schedules and feedback loops from agent failure modes back to expert panels."
sources: ["agentic"]
sourceVaultSlug: "hbr-seg-agentic"
originDay: 6
articleStem: "hbr-new-27-teach-ai-your-decisions"
sourceUrl: "https://hbr.org/2026/06/teach-your-ai-how-you-make-decisions"
sourceTitle: "Teach Your AI How You Make Decisions"
---
# How is codified judgment maintained as the business evolves?

**Open question:** The article notes that agent behavior must be "continually refined as the work, the business, and the judgment around it evolve," but it does not provide a specific framework for version control, auditing, or updating the context files/transcripts once the initial [[concept-codifying-judgment|extraction]] is complete.

**Possible resolution path:** Develop a lifecycle management framework for "judgment files" — regular audit schedules and feedback loops that route agent failure modes back to expert panels ([[action-convene-expert-panels]]). This connects to the risk-framework integration argued in [[cp-compliance-risk-frameworks]] and the maintainability warning in [[cp-sops-still-valuable]].