1、AI Agents in Action:A Playbook for Trusted Adoption,Authorization and Scaling I N S I G H T R E P O R TM A Y 2 0 2 6In collaboration withCapgeminiImages:Getty ImagesDisclaimer This document is published by the World Economic Forum as a contribution to a project,insight area or interaction.The findin
2、gs,interpretations and conclusions expressed herein are a result of a collaborative process facilitated and endorsed by the World Economic Forum but whose results do not necessarily represent the views of the World Economic Forum,nor the entirety of its Members,Partners or other stakeholders.2026 Wo
3、rld Economic Forum.All rights reserved.No part of this publication may be reproduced or transmitted in any form or by any means,including photocopying and recording,or by any information storage and retrieval system.ContentsForeword 3Executive summary 4Introduction 51 Agent guidelines 71.1 Establish
4、ing a shared language for autonomy,8 authority and consequence1.2 Allocating decision rights and accountability across the life cycle 91.3 Defining when agentic systems are the appropriate solution 91.4 Sequencing adoption and prioritizing early use cases 101.5 Deployment contexts and baseline gover
5、nance 111.6 Defining the humanagent operating model 121.7 From enterprise guidelines to deployment authorization 122 ACAP:The Agent Capability and Authorization Profile 142.1 ACAP structure and how to use it 162.2 System design and assessment 172.3 Prepare and deploy 232.4 Monitor and improve 27Conc
6、lusion 30Appendix:ACAP summary playbook 31Contributors 34Endnotes 37AI Agents in Action2ForewordArtificial intelligence(AI)agents have left research laboratories to become a permanent fixture in organizational workflows.This shift has outpaced our existing governance frameworks.Where the first publi