Back to the AI Control Library
12 controls | 12 core | 0 supplemental
Why this family matters
People need to know when they are dealing with AI, what it can reasonably do, and where its output needs scrutiny. This family addresses the information required for users, operators and affected people to understand an AI system's role and limitations. Transparency must be useful to the audience receiving it.
What this family covers
The controls cover explainability requirements and evaluation, AI identity disclosure, capability and limitation notices, source attribution, uncertainty, oversight transparency and explanation rights. They also address transparency expectations for systems used in higher-impact settings.
How the four control types apply
Contextual controls shape how users interpret and rely on AI through disclosure, explanations and information about uncertainty. Operational controls establish the processes for producing and reviewing that information. Technical mechanisms support delivery and traceability, while Governance controls establish applicable external disclosure obligations.
Where to begin
Identify each audience and the decisions it must make, then provide clear information about AI involvement, limitations and available oversight.
Controls in this family
This family contains 12 controls: 12 core and 0 supplemental. Each page provides requirements, implementation steps, verification criteria, evidence, escalation and a worked example. Applicability depends on the system, information and risks involved.
Core controls
- CF06-001: Explainability Requirements Definition
- CF06-002: Explainability Evaluation & Documentation
- CF06-003: AI System Transparency Obligations
- CF06-004: AI Identity Disclosure
- CF06-005: AI Capability & Limitation Notices
- CF06-006: AI Source Citation & Attribution
- CF06-007: Organizational AI Transparency Policy
- CF06-008: Human Oversight Transparency
- CF06-009: AI Output Uncertainty & Confidence Disclosure
- CF06-010: High-Risk AI System Transparency Requirements
- CF06-011: Automated Decision Explanation Rights
- CF06-012: AI System Public Disclosure
Applying this family
Define the audience and decision before choosing an explanation or disclosure. Use supported facts, tested capability claims and usable contact routes. Distinguish system-level information, explanation of a specific outcome and evidence of human involvement. Revisit communications when the underlying system or operating arrangements change.
Version 1.0