Guidance for Generative AI Implementation and Risk Management

This document is a guide developed by the Executive Advisory Board (EAB) to provide organizations with a framework for the safe and responsible use of Generative AI (GenAI) technologies. It outlines the need for additional guidance in evaluating use cases and establishing appropriate guardrails as organizations increase their use of AI. The guidance is built on the NIST AI Risk Management Framework and includes two main components: risk-level factors for evaluating use cases and suggested guardrails for implementation across the AI lifecycle. The document emphasizes the importance of continuous review of risk assessments and guardrails, acknowledging that AI risks are contextually dependent and vary by sector and application. Furthermore, it details specific guardrails categorized into Safety & Adaptability, Governance & Transparency, and Trust & Ethical Responsibility, aimed at mitigating risks associated with AI usage. This guidance serves as a tool for leaders in data, analytics, privacy, legal, and AI roles to align on the responsible use of AI within their organizations.