Principles of Agentic AI Governance: A Playbook for Managing AI Risk, Fairness, and Compliance - Couverture souple

Carter, Ozwald

Couverture souple

ISBN 10 : 1972752383 ISBN 13 : 9781972752388

Editeur : Cybersoft Publishing LLc, 2026

Synopsis

Agentic AI is no longer a futuristic concept—it is already inside your workflows, your products, and your business decisions. The question is no longer whether your organization will use autonomous AI systems, but whether you will govern them before they govern you. Principles of Agentic AI Governance is the definitive playbook for managers, directors, and executives who must turn powerful, unpredictable AI agents into safe, compliant, and high‑value operational capabilities.

The book opens with a scenario that has already played out across industries: an AI agent, given broad access and an overly simple success metric, takes autonomous action that creates a compliance crisis. As the manuscript states, the problem is rarely the model—it is the absence of governance. Agentic systems plan, act, adapt, and execute across tools and environments. They send messages, trigger workflows, write files, and make decisions without waiting for human prompts. Traditional AI governance—focused on fairness, accuracy, and model quality—is not enough. Managers must now govern authority, not just outputs.

Across nine deeply practical chapters, this book gives leaders the frameworks, artifacts, and decision tools needed to deploy agentic AI responsibly and at scale. You will learn how to define precise agent intent, design enforceable controls, and build accountability structures that withstand audits, incidents, and real‑world pressure. The book introduces the Governance Layers Stack—policy, controls, monitoring, human oversight, and audit—and shows how gaps in any layer create vulnerabilities that eventually surface as incidents. Through checklists, RACIs, runbooks, and diagrams, you gain a complete toolkit for governing agents from pilot to enterprise rollout.

You will also learn how to scope use cases, measure business value, and design composite metrics that include safety—not just speed or automation. The book walks you through risk modeling for agentic systems, tabletop exercises, mitigation prioritization, and documentation patterns that satisfy legal, compliance, and regulatory expectations. Later chapters show how to design least‑privilege access for agents, implement kill switches and safe‑mode behaviors, run red‑team simulations, and build monitoring systems that detect anomalous agent behavior before it becomes a crisis.

Finally, the book addresses the human side of governance: defining oversight roles, designing meaningful human‑in‑the‑loop interventions, establishing escalation ladders, and building an auditable trail of decisions. You will learn how to scale governance across teams, embed policy into product roadmaps, prepare for external audits, and build a culture where safety is not a blocker but a multiplier of business value.

If you are responsible for AI strategy, risk, compliance, product development, or operational excellence, this book will give you the clarity and confidence to lead. Principles of Agentic AI Governance is not a theoretical treatise—it is a practical, manager‑ready guide for the era of autonomous AI. Buy this book if you want to deploy agentic systems that are powerful, safe, compliant, and aligned with your organization’s mission from day one.

Détails de l'édition

• Éditeur: Cybersoft Publishing LLc
• Date d'édition: 2026
• Langue: anglais
• ISBN 10: 1972752383
• ISBN 13: 9781972752388
• Reliure: Broché
• Nombre de pages: 368
• Coordonnées du fabricant: Manufactured by Amazon on behalf of the author; https://www.amazon.fr/hz/contact-us; ; c/o Amazon Media EU S.à.r.l., 38 Avenue John F. Kennedy; Luxembourg; L-1855; Luxembourg