Ensure the most-advanced AI systems are implemented in a responsible way that furthers innovation and prevents risks with this actionable guide.
Safer Agentic AI provides senior professionals with the skills to ensure the safe and beneficial use of advanced AI systems. Agentic and advanced AI systems represent a tremendous opportunity for business to improve their processes and increase productivity. However, these systems come with critical risks that organizations will need to manage. This book will show how to harness the power of these systems in a responsible and safe way.
Covering how to steer the future of AI responsibly and avoid potential pitfalls, this book will explain how organizations can use agentic AI in ways that safeguard their values and future. It draws on technical insights and ethical considerations to offer practical governance frameworks that can work in any organization. Written by two experts in AI ethics and safety, this book will cover key areas such as goal alignment, value alignment, transparency, governance and the challenges of maintaining human control over increasingly autonomous systems.
Rezensionen / Stimmen
"Nell Watson and Ali Hessami's new book is just in time, offering the actionable guidance we must have to ensure AI truly aligns with human values." * David D Luxton, Affiliate Professor, University of Washington School of Medicine * "This book is an essential read for those seeking to safeguard humanity's future in the age of AI, serving as a reminder that we all are stewards and have a significant role to play. The book provides practical action items for agentic AI safety, combining ethical insight with governance strategies to help organizations harness agentic AI responsibly and effectively." * Keeley A Crockett, Professor in Computational Intelligence, Manchester Metropolitan University * "With clarity and courage, Nell Watson and Ali Hessami expose how AI shapes our minds, organizations and planet and how we must build the guardrails to keep it serving life, not running it." * Thijs Pepping, philosopher of technology, SogetiLabs and co-author of Real Fake * "Safer Agentic AI advances a pragmatic control architecture by regulating agents by what they can actually do. Imposing measurable safety gates on plans, tools, and autonomy, Nell Watson and Ali Hessami turn alignment from aspiration into auditable engineering." * Roman Yampolskiy, author of AI: Unexplainable, Unpredictable, Uncontrollable * "Agentic AI can bring humanity into a better future - if it is governed responsibly. I can think of no one better than Nell Watson and Ali Hessami to guide us all on that crucial journey." * Kris Ostergaard, author of Transforming Legacy Organizations * "Nell Watson and Ali Hessami have produced a guide for safely and ethically engaging with agentic AI. This work presents clear descriptions of many of the ways in which human and computational intelligences interact. The authors' approach performs a vital service for both information professionals and concerned citizens alike when they discuss these once-obscure and subtle aspects of intelligence and interactions. In doing so, they reveal points of intervention to help humans and organizations to more successfully, safely and ethically engage with computationally intelligent systems." * Scott L David, Executive Director, Information Risk and Synthetic Intelligence Research Initiative, University of Washington * "Nell Watson and Ali Hessami address the potential risks of agentic AI. The authors provide an overview of the concepts and principles of this fast-moving technology and consider the key issues, such as how to align the behaviour of agents with human goals and values. They specifically address the topics of safety, security and ethics in the context of AI and other important aspects of controlling and governing agentic AI. The work is timely and thought provoking; without the approaches they advocate there is a risk that the development of agentic AI will outpace the community's ability to ensure positive outcomes from its use." * John A McDermid, Lloyd's Register Foundation Chair in Safety, University of York *
Sprache
Verlagsort
Zielgruppe
Für höhere Schule und Studium
Für Beruf und Forschung
Produkt-Hinweis
Maße
Höhe: 234 mm
Breite: 156 mm
ISBN-13
978-1-3986-2545-7 (9781398625457)
Copyright in bibliographic data and cover images is held by Nielsen Book Services Limited or by the publishers or by their respective licensors: all rights reserved.
Schweitzer Klassifikation
Nell Watson is a researcher and applied tech ethicist. She advises organizations on their machine learning strategies, specializing in AI research and ethical advocacy. She is President of the European Responsible AI Office and the author of Taming the Machine published by Kogan Page. She lives in Belfast, UK.
Ali Hessami is a systems engineer specializing in risk analysis, assessment and safety. He is Vice Chair of the IEEE AI Ethics Certification Programme and Chair of the IEEE 7000 AI Ethics Standard. He lives in London, UK.
Chapter - 00: Introduction
Section - ONE: The Dawn of Agency;
Chapter - 01: AI Overview;
Chapter - 02: The Age of Agentic AI;
Section - TWO: The Core Alignment Problem
Chapter - 03: Goal Alignment of Agentic AI Systems;
Chapter - 04: Value Alignment and Embedding Ethics;
Chapter - 05: Deceptive and Power-Seeking AI;
Chapter - 06: Utility Convergence;
Section - THREE: The Governance and Safety Toolkit
Chapter - 07: Safety Fundamentals;
Chapter - 08: Security and Guardrails;
Chapter - 09: Agentic Conflict
Chapter - 10: Transparency and Interpretability;
Chapter - 11: AI Governance;
Chapter - 12: Weighted Capability Governance;
Section - FOUR: The Human and Societal Interface
Chapter - 13: Working with Agentic AI;
Chapter - 14: Psychosecurity;
Chapter - 15: Who Goes There?;
Chapter - 16: Self-Watching Machines;
Chapter - 17: Sustainability and Resource Stewardship;
Section - FIVE: The Unchartered Frontier
Chapter - 18: Disruptive Waves;
Chapter - 19: Frontier Concerns;
Chapter - 20: Superintelligence Strategy;
Chapter - 21: Ten Principles for a Positive Future;
Chapter - 22: Further Resources;
Chapter - 23: Glossary
