The Insider You Built: Accountability and Governance of Autonomous AI Agents
暫譯: 你打造的內部人員：自主 AI Agent 的問責與治理

An operational playbook for maintaining control as AI agents scale across the enterprise

AI agents are rapidly becoming embedded across enterprise systems. They schedule meetings, generate reports, approve transactions, interact with customers, and make decisions across digital environments. Organizations have, in effect, created a new kind of insider: software trusted to act on the company's behalf.

Seventy-nine percent of executives report that AI agents are already deployed in their organizations, yet few companies have governance frameworks for systems that can plan actions and operate across the business with minimal human direction. As these systems grow more capable and integrated, governing them effectively is emerging as a defining leadership challenge of the AI era.

In The Insider You Built: How Organizations Stay in Control of Autonomous AI Agents, Camille Stewart Gloster introduces the Agent-Centered Enforcement and Attribution (ACE) Framework, a practical model for investigating, attributing, and responding to failures when autonomous AI systems act with delegated authority. The book reframes AI incidents as governance challenges, helping organizations identify breakdowns across human and machine decision paths and respond when AI creates operational, regulatory, or reputational risk. The ACE Framework provides durable governance infrastructure that evolves alongside AI capabilities, enabling leaders to maintain control as adoption accelerates.

Readers will discover how to:

• Adapt insider risk, security, and governance programs for autonomous AI agents operating with delegated authority
• Establish board- and executive-level oversight that connects AI governance to enterprise risk, performance, and regulatory obligations
• Investigate AI-driven incidents and attribute responsibility across complex human-machine decision paths
• Coordinate responses across security, legal, and leadership teams

Through operational playbooks, governance models, and real-world response strategies, this book equips leaders to manage the risks introduced when AI agents operate across enterprise systems while enabling responsible innovation and performance.