Who this book is for

As we continue to work with enterprises, advising and guiding them as they seek to transform themselves to become data-driven – producing their own actionable insights, machine-learning models, and AI at scale – we are acutely aware of their concerns and questions regarding AI assurance.

This book is written for a wide range of professionals in the field of enterprise AI and machine learning. This includes data scientists, machine learning engineers, AI practitioners, IT professionals, business stakeholders, software engineers, AI ethicists, and, last but not least, enterprise change leaders. These are the people working within the enterprise to both affect the changes required to become data-driven and to successfully develop and deliver AI models at scale.

The book covers a comprehensive range of topics, from XAI and ethical considerations to model governance and compliance standards, and provides practical guidance on using tools such as hyperscalers, open source tools, and Microsoft Fairlearn. It is a valuable resource for those who are interested in understanding the latest developments in AI governance, including the role of internal AI boards, the importance of data governance, and the latest industry standards and regulations.

The book is also relevant for AI professionals in a variety of industries, including healthcare, customer service, and finance, using conversational AI and predictive analytics. Whether you are a business stakeholder responsible for making decisions about AI adoption, an AI ethicist concerned with the ethical implications of AI, or an AI practitioner responsible for building and deploying models, this book provides valuable insights and practical guidance on building responsible and transparent AI models.

Essential chapters tailored to distinct AI-related positions

For AI ethicists, auditors, and compliance personnel, the most relevant chapters are as follows:

• Chapter 1, A Explainable and Ethical AI Primer
• Chapter 5, Model Governance, Audit, and Compliance
• Chapter 6, Enterprise Starter Kit for Fairness, Accountability, and Transparency
• Chapter 10, Foundational Models and Azure OpenAI

These chapters focus on explainable and ethical AI, model governance, compliance standards, responsible AI implementation, and the challenges associated with large language models.

Managers and business stakeholders will find the following chapters most relevant:

• Chapter 2, Algorithms Gone Wild
• Chapter 5, Model Governance, Audit, and Compliance
• Chapter 6, Enterprise Starter Kit for Fairness, Accountability, and Transparency

These chapters cover the impact of bias in AI, the importance of transparency and accountability in AI-driven decision-making, and the practical aspects of implementing AI governance within an organization.

Data scientists and machine learning engineers will find the entire book quite useful, but the most relevant chapters for data scientists and machine learning engineers are as follows:

• Chapter 1, Explainable and Ethical AI Primer
• Chapter 3, Opening the Algorithmic Black Box
• Chapter 4, Robust ML - Monitoring and Management
• Chapter 7, Interpretability Toolkits and Fairness Measures - AWS, GCP, Azure, and AIF 360
• Chapter 8, Fairness in AI Systems with Microsoft Fairlearn
• Chapter 9, Fairness Assessment and Bias Mitigation with Fairlearn and the Responsible AI Toolbox

These chapters provide valuable information on explainable and ethical AI, model interpretability, monitoring model performance, and practical applications of fairness and bias mitigation techniques.

While the book covers advanced-level concepts, it is written in an accessible style and assumes a basic understanding of AI and machine learning concepts. However, those with less experience may need to put in additional effort to fully understand the material.