AI Organizational Responsibilities - Core Security Responsibilities
Released: 05/05/2024
This publication from the CSA AI Organizational Responsibilities Working Group provides a blueprint for enterprises to fulfill their core information security responsibilities pertaining to the development and deployment of Artificial Intelligence (AI) and Machine Learning (ML). Expert-recommended best practices and standards, including NIST AI RMF, NIST SSDF, NIST 800-53, and CSA CCM, are synthesized into 3 core security areas: data protection mechanisms, model security, and vulnerability management. Each responsibility is analyzed using quantifiable evaluation criteria, the RACI model for role definitions, high-level implementation strategies, continuous monitoring and reporting mechanisms, access control mapping, and adherence to foundational guardrails.
Key Takeaways:
- The components of the AI Shared Responsibility Model
- How to ensure the security and privacy of AI training data
- The significance of AI model security, including access controls, secure runtime environments, vulnerability and patch management, and MLOps pipeline security
- The significance of AI vulnerability management, including AI/ML asset inventory, continuous vulnerability scanning, risk-based prioritization, and remediation tracking
The other two publications in this series discuss the AI regulatory environment and a benchmarking model for AI resilience. By outlining recommendations across these key areas of security and compliance in 3 targeted publications, this series guides enterprises to fulfill their obligations for responsible and secure AI development and deployment.
Topics:
Best For:
- CISOs and Chief AI Officers
- Business leaders, decision makers, and shareholders
- AI engineers, analysts, and developers
- Policymakers and regulators
- Customers and the general public
Premier AI Safety Ambassadors
Premier AI Safety Ambassadors play a leading role in promoting AI safety within their organization, advocating for responsible AI practices and promoting pragmatic solutions to manage AI risks. Learn more about how your organization could participate and take a seat at the forefront of AI safety best practices.
