Responsible AI: Best Practices for Creating Trustworthy AI Systems

Dr. Qinghua Lu is a principal research scientist and leads the Responsible AI science team at CSIROs Data61. She received her PhD from University of New South Wales in 2013. Her current research interests include responsible AI, software engineering for AI/GAI, and software architecture. She has published 150+ papers in premier international journals and conferences. Her recent paper titled Towards a Roadmap on Software Engineering for Responsible AI received the ACM Distinguished Paper Award. Dr. Lu is part of the OECD.AIs trustworthy AI metrics project team. She also serves a member of Australias National AI Centre Responsible AI at Scale think tank. She is the winner of the 2023 APAC Women in AI Trailblazer Award.
 
Dr./Prof. Liming Zhu is a Research Director at CSIROs Data61 and a conjoint full professor at the University of New South Wales (UNSW). He is the chairperson of Standards Australias blockchain committee and contributes to the AI trustworthiness committee. He is a member of the OECD.AI expert group on AI Risks and Accountability, as well as a member of the Responsible AI at Scale think tank at Australias National AI Centre. His research program innovates in the areas of AI/ML systems, responsible/ethical AI, software engineering, blockchain, regulation technology, quantum software, privacy, and cybersecurity. He has published more than 300 papers on software architecture, blockchain, governance and responsible AI. He delivered the keynote Software Engineering as the Linchpin of Responsible AI at the International Conference on Software Engineering (ICSE) 2023.
 
Prof. Jon Whittle is Director at CSIROs Data61, Australias national centre for R&D in data science and digital technologies. With around 850 staff and affiliates, Data61 is one of the largest collections of R&D expertise in Artificial Intelligence and Data Science in the world. Data61 partners with more than 200 industry and government organisations, more than 30 universities, and works across vertical sectors in manufacturing, health, agriculture, and the environment. Prior to joining Data61, Jon was Dean of the Faculty of Information Technology at Monash University.
 
Dr. Xiwei Xu is a principal research scientist and the group leader of the software systems research group at Data61, CSIRO. With a specialization in software architecture and system design, she is  at the forefront of research in these fields. Xiwei is identified by the Bibliometric Assessment of Software Engineering Scholars and Institutions as a top scholar and ranked 4th in the world (20132020) as the most impactful SE researchers by JSS (Journal of Systems and Software), a well-recognized academic journal in software engineering research.

    Preface.. . . . . . . . . . . . . . . . . xv
    About the Author.. . . . . . . . . . . . . . xix
Part I Background and Introduction. . . . . . . . . . . . .1
1 Introduction to Responsible AI. . . . . . . . . 3
    What Is Responsible AI?. . . . . . . . . . . . 4
    What Is AI?. . . . . . . . . . . . . . 6
    Developing AI Responsibly: Who Is Responsible for Putting the
    Responsible into AI?.. . . . . . . . . . . . 8
    About This Book.. . . . . . . . . . . . . 9
    How to Read This Book.. . . . . . . . . . . . 11
2 Operationalizing Responsible AI: A Thought ExperimentRobbie the Robot.. . . . . . . . 13
    A Thought ExperimentRobbie the Robot.. . . . . . . . 13
    Summary. . . . . . . . . . . . . . 22
Part II Responsible AI Pattern Catalogue. . . . . . . . . . . 23
3 Overview of the Responsible AI Pattern Catalogue. . . . . 25
    The Key Concepts.. . . . . . . . . . . . . 25
    Why Is Responsible AI Different?. . . . . . . . . . 30
    A Pattern-Oriented Approach for Responsible AI.. . . . . . . 32
4 Multi-Level Governance Patterns for Responsible AI.. . . . 39
    Industry-Level Governance Patterns. . . . . . . . . 42
    Organization-Level Governance Patterns.. . . . . . . . 56
    Team-Level Governance Patterns.. . . . . . . . . . 72
    Summary. . . . . . . . . . . . . . 85
5 Process Patterns for Trustworthy Development Processes. . . 87
    Requirements.. . . . . . . . . . . . . 88
    Design. . . . . . . . . . . . . . . 96
    Implementation.. . . . . . . . . . . . . 105
    Testing. . . . . . . . . . . . . . . 110
    Operations. . . . . . . . . . . . . . 114
    Summary. . . . . . . . . . . . . . 120
6 Product Patterns for Responsible-AI-by-Design.. . . . . 121
    Product Pattern Collection Overview.. . . . . . . . . 122
    Supply Chain Patterns. . . . . . . . . . . . 123
    System Patterns. . . . . . . . . . . . . 134
    Operation Infrastructure Patterns. . . . . . . . . 141
    Summary. . . . . . . . . . . . . . 158
7 Pattern-Oriented Reference Architecture for Responsible-AI-by-Design. . . . . . . . . 159
    Architectural Principles for Designing AI Systems. . . . . . 160
    Pattern-Oriented Reference Architecture.. . . . . . . . 161
    Summary. . . . . . . . . . . . . . 165
8 Principle-Specific Techniques for Responsible AI.. . . . . 167
    Fairness.. . . . . . . . . . . . . . 167
    Privacy. . . . . . . . . . . . . . . 172
    Explainability. . . . . . . . . . . . . 178
    Summary. . . . . . . . . . . . . . 182
Part III Case Studies. . . . . . . . . . . . . . .  183
9 Risk-Based AI Governance in Telstra. . . . . . . 185
    Policy and Awareness.. . . . . . . . . . . . 186
    Assessing Risk.. . . . . . . . . . . . . 188
    Learnings from Practice. . . . . . . . . . . 192
    Future Work. . . . . . . . . . . . . . 195
10 Reejig: The Worlds First Independently Audited Ethical Talent AI.. . . . . . . . . . . 197
    How Is AI Being Used in Talent?.. . . . . . . . . . 198
    What Does Bias in Talent AI Look Like?.. . . . . . . . 200
    Regulating Talent AI Is a Global Issue.. . . . . . . . . 201
    Reejigs Approach to Ethical Talent AI. . . . . . . . . 202
    How Ethical AI Evaluation Is Done: A Case Study in Reejigs World-First Independently Audited Ethical Talent AI. . . . . . . . 204
    Overview.. . . . . . . . . . . . . 204
    Project Overview. . . . . . . . . . . . . 206
    The Ethical AI Framework Used for the Audit.. . . . . . . 207
    The Benefits of Ethical Talent AI.. . . . . . . . . . 210
    Reejigs Outlook on the Future of Ethical Talent AI.. . . . . . 211
11 Diversity and Inclusion in Artificial Intelligence.. . . . . 213
    Importance of Diversity and Inclusion in AI.. . . . . . . 215
    Definition of Diversity and Inclusion in Artificial Intelligence. . . . 216
    Guidelines for Diversity and Inclusion in Artificial Intelligence. . . . 219
    Conclusion.. . . . . . . . . . . . . . 234
Part IV Looking to the Future. . . . . . . . . . . . . 237
12 The Future of Responsible AI.. . . . . . . . . 239
    Regulation. . . . . . . . . . . . . . 241
    Education.. . . . . . . . . . . . . . 242
    Standards.. . . . . . . . . . . . . . 244
    Tools.. . . . . . . . . . . . . . . 245
    Public Awareness.. . . . . . . . . . . . 246
    Final Remarks.. . . . . . . . . . . . . 246
Part V Appendix. . . . . . . . . . . . . . . . 249
 
9780138073923, TOC, 11/7/2023