AI governance and risk management frameworks provide structured approaches for organizations to develop, deploy, and monitor artificial intelligence systems responsibly while meeting regulatory obligations. As AI moves from experimentation into production-critical systems in 2026, the convergence of NIST guidance, EU legislation, and state-level US regulations creates enforceable accountability across the entire AI lifecycle. Understanding risk classification tiers, implementing continuous monitoring infrastructure, and establishing cross-functional governance bodies are no longer optional—they are foundational requirements for any organization using AI in high-stakes domains such as hiring, healthcare, law enforcement, and financial services.
What This Cheat Sheet Covers
This topic spans 30 focused tables and 189 indexed concepts, 127 flashcards. Below is a complete table-by-table outline of this topic, spanning foundational concepts through advanced details.
A jump-to index of every table row in this cheat sheet.
An interactive map of every table and concept in this topic.
Table 1: Core Governance Frameworks
Organizations align their AI programs to established frameworks that provide structured guidance for risk identification, assessment, and mitigation across the AI lifecycle.
| Framework | Example | Description | |
|---|---|---|---|
Four functions: Govern, Map, Measure, Manage | • Voluntary US framework organizing AI risk activities into four interconnected functions • widely adopted across sectors and designed to improve AI trustworthiness without mandating specific technical solutions | ||
Risk-based tiers: Unacceptable, High, Limited, Minimal | • World's first comprehensive AI law classifying systems into risk categories with corresponding obligations • enforceable from August 2, 2026, with fines up to €35M or 7% of global turnover for prohibited practices | ||
AI management system standard | • First internationally certifiable standard for AI management systems • defines requirements for establishing, implementing, and maintaining AI governance aligned with risk-based principles and continuous improvement | ||
Five values-based principles | • Foundational principles promoting innovative, trustworthy AI that respects human rights and democratic values • adopted globally and referenced by multiple national frameworks | ||
193 member states commitment | • First global standard-setting instrument on AI ethics • provides detailed policy guidance on proportionality, safety, fairness, and sustainability in AI development | ||
High-risk system protections | • US state-level law requiring reasonable care to protect consumers from algorithmic discrimination in high-risk AI systems • effective February 2026 with impact assessment obligations | ||
Training data disclosure requirements | • Requires covered providers to publicly document training datasets and contractual opt-out mechanisms • effective January 1, 2026, as part of California's broader AI regulatory push |