Purpose: Define who has authority to make gray-zone judgment calls for AI systems—eliminating the responsibility vacuum where everyone sees the problem but no one can own the solution.

How to use: Duplicate this page and customize for your organization. Fill in the templates below for each type of gray-zone decision your AI systems will generate.

🎯 Why This Framework Exists

Traditional governance assigns clear ownership for binary decisions:

• Legal approves or rejects contracts
• Finance approves budgets within thresholds or flags overages
• Security passes or fails architecture reviews

Gray-zone AI systems break this model. Decisions aren't binary:

• Is 87% accuracy good enough? (Depends on context)
• Is 8% demographic variance acceptable? (Depends on legal/ethical standards)
• Does 5% drift require retraining? (Depends on criticality and resources)

Without explicit authority for gray-zone judgment calls, you get:

• Paralysis (everyone waits for someone else to decide)
• Escalation overload (everything goes to executives)
• Shadow decisions (people act without authority and hope no one notices)

This framework prevents that by assigning:

• Primary authority: Who makes the call
• Required consultation: Who must be consulted before the call is made
• Veto rights: Who can override if thresholds are crossed