Every reviewed AI governance resource in one place — reports, standards, guidance and regulation.
A lifecycle-driven framework offering 70+ mapped AI-specific risks, actionable safeguards, and crosswalks to ISO 42001, NIST AI RMF, OWASP Top 10 for LLMs, and more – aimed at embedding security into every phase of AI development and deployment
A practical, product-integrated framework for managing AI risks across the ML lifecycle—rooted in Databricks’ tooling and aligned with enterprise data governance priorities.
A structured guide to identifying and mitigating privacy risks in LLMs—covering data leakage, user inference, training data exposure, and strategies for auditability and control.
A detailed legislative framework proposing oversight, obligations, and incentives for AI systems—covering everything from foundation models and open-source exemptions to risk-based licensing and whistleblower protection.
Customers acknowledge the need to secure AI systems but simply do not know how.
Organizations that measure the value of AI ethics could be a step ahead. Our holistic AI ethics framework considers three types of ROI.
Without a shared understanding of how bias enters and operates in AI systems, law enforcement agencies risk embedding discrimination into everyday operations.
Frontline users will need a high degree of discretion over how they use AI assistants. But this must be matched with rigorous oversight and clear internal boundaries.
AIME distils key principles from existing AI regulations, standards and frameworks to provide an accessible starting point for organisations to assess and improve their AI management systems.
The Agentic Oversight Framework ensures agents are contained and embedded into a secure environment that meets institutional requirements for data handling, oversight, and auditability.
Risk tiers clarify the harms AI might present and identify the measures being taken to prevent them.
AI Governance by Design (AIGD) integrates ethical, legal, and societal considerations directly into AI system development from inception.
We are releasing our Impact Assessment Template externally to share what we have learned, invite feedback from others, and contribute to the discussion about building better norms and practices around AI.
“Organizations might not even have clear answers to more fundamental questions such as: how does the output of an AI system affect things in practice? Or even—is AI used at all here?”
“If [AI agents] live up to their promise, they may soon become highly capable digital coworkers or personal assistants… moving AI ‘through the chat window and into the real world.’” — p. 7
Public institutions must not merely invite public comment—they must share power. Communities must be resourced and authorized to co-govern the design, deployment, and evaluation of AI systems.
This practical and comprehensive implementation guide offers a structured, checklist-based framework for organizations seeking to operationalize AI data governance. Rooted
This report argues that traditional, consent-based models for governing AI data reuse are no longer adequate—especially in low- and middle-income countries (LMICs), where power asymmetries and exploitative data practices are common.
This document introduces the ISO/IEC 42005:2025 standard and offers a detailed, visually engaging explanation of how it enables standardized, interoperable AI impact assessments. Written and compiled by Georg Philip Krog.
This report dissects the generative AI ecosystem through the lens of human rights. Instead of focusing only on the models or outputs, it zooms out to examine the entire value chain—covering everything from raw compute and data to the deployment of tools in the market.
The Agentic AI Red Teaming Guide is the most comprehensive hands-on security testing manual yet for red teaming autonomous AI systems. Developed by CSA and OWASP with input from over 50 contributors.
This guided template helps public and private sector leaders build structured, role-based AI literacy programs aligned with legal obligations (like Article 4 of the EU AI Act) and governance best practices.
This policy paper zeroes in on the governance challenges posed by increasingly autonomous AI agents—systems capable of initiating and managing multi-step tasks with minimal human intervention.
The Cyber Governance Code of Practice is a government-endorsed framework aimed at medium and large organisations, designed to help boards take ownership of cybersecurity risks. It focuses on aligning cybersecurity with strategic, operational, and cultural goals of the business.
Curated Library of AI Governance Resources
