Description
Book SynopsisSo far, little effort has been devoted to developing practical approaches on how to develop and deploy AI systems that meet certain standards and principles. This is despite the importance of principles such as privacy, fairness, and social equality taking centre stage in discussions around AI. However, for an organization, failing to meet those standards can give rise to significant lost opportunities. It may further lead to an organization''s demise, as the example of Cambridge Analytica demonstrates. It is, however, possible to pursue a practical approach for the design, development, and deployment of sustainable AI systems that incorporates both business and human values and principles.
This book discusses the concept of sustainability in the context of artificial intelligence. In order to help businesses achieve this objective, the author introduces the sustainable artificial intelligence framework (SAIF), designed as a reference guide in the development and deployment
Table of Contents● Chapter 1: AI in our Society● Chapter goal: Reviews the place of AI within our society, discuss the various challenges that it AI faces, and introduces the foundational concepts of our sustainable AI framework ○ 1.1 The Need for Artificial Intelligence○ 1.2 Challenges of Artificial Intelligence○ 1.3 Sustainable Artificial Intelligence
● Chapter 2 Ethics of the Data Science Practice● Chapter goal: Reviews the human factor pillar of artificial intelligence, the relevance of ethics in AI and the source of ethical hazards in AI ○ 2.1 Introduction○ 2.2 Ethics and their relevance to AI○ 2.3 Ethical nature of AI inferencing capability○ 2.4 Data – The business asset○ 2.5 AI regulatory outlook○ 2.6 Conclusion
● Chapter 3 Overview of the Sustainable Artificial Intelligence Framework (SAIF)● Chapter goal: Summarises the SAIF framework for the development and deployment of AI applications
● Chapter 4 Intra-organizational understanding of AI: Towards Transparency● Chapter goal: Discusses the need for understanding AI at the organization’s level and introduces concepts of AI governance○ 4.1 Introduction○ 4.2 Data Science Development Process○ 4.3 AI development process Controls○ 4.4 Governance■ 4.4.1 Expectations from AI governance■ 4.4.2 People and Values■ 4.4.3 Assessment of AI governance arrangements○ 4.5 Conclusion
● Chapter 5 AI Performance Measurement: Think business values and objectives● Chapter goal: Summarises performance metrics for evaluating AI systems and introduces a framework to account for the human factor of AI○ 5.1 Introduction○ 5.2 AI performance metrics overview■ 5.2.1 Supervised problems ■ 5.2.2 Unsupervised problems ○ 5.3 Beyond traditional AI performance metrics■ 5.3.1 Soft performance metrics■ 5.3.2 From AI performance metrics to business objectives○ 5.4 Conclusion
● Chapter 6 SAIF in Action● Chapter goal: This chapter illustrates how SAIF would work in practice through use cases
● Chapter 7 Alternatives avenues for regulating AI systems● Chapter goal: Draws from experiences in academic, Telecom/Utility, and healthcare sectors to explore and examine the need for industry specific regulations.
● Chapter 8 AI decision-making – from expectations to reality: The use case of healthcare● Chapter goal: Explores the use of artificial intelligence in the healthcare, its practical limitations an implications
● Chapter 9 Conclusions and discussion● Chapter goal: Presents concluding remarks and discuss current lack of standards ○ 9.1 Conclusions○ 9.2 Need for standards and definitions
