Artificial Intelligence for Business

Artificial Intelligence for Business: A Roadmap for Getting Started with AI will provide the reader with an easy to understand roadmap for how to take an organization through the adoption of AI technology. It will first help with the identification of which business problems and opportunities are right for AI and how to prioritize them to maximize the likelihood of success. Specific methodologies are introduced to help with finding critical training data within an organization and how to fill data gaps if they exist. With data in hand, a scoped prototype can be built to limit risk and provide tangible value to the organization as a whole to justify further investment. Finally, a production level AI system can be developed with best practices to ensure quality with not only the application code, but also the AI models. Finally, with this particular AI adoption journey at an end, the authors will show that there is additional value to be gained by iterating on this AI adoption lifecycle and improving other parts of the organization.

Preface ix

Acknowledgments xi

Chapter 1 Introduction 1

Case Study #1: FANUC Corporation 2

Case Study #2: H&R Block 4

Case Study #3: BlackRock, Inc. 5

How to Get Started 6

The Road Ahead 10

Notes 11

Chapter 2 Ideation 13

An Artificial Intelligence Primer 13

Becoming an Innovation-Focused Organization 23

Idea Bank 25

Business Process Mapping 27

Flowcharts, SOPs, and You 28

Information Flows 29

Coming Up with Ideas 31

Value Analysis 31

Sorting and Filtering 34

Ranking, Categorizing, and Classifying 35

Reviewing the Idea Bank 37

Brainstorming and Chance Encounters 38

AI Limitations 41

Pitfalls 44

Action Checklist 45

Notes 46

Chapter 3 Defining the Project 47

The What, Why, and How of a Project Plan 48

The Components of a Project Plan 49

Approaches to Break Down a Project 53

Project Measurability 62

Balanced Scorecard 63

Building an AI Project Plan 64

Pitfalls 66

Action Checklist 69

Chapter 4 Data Curation and Governance 71

Data Collection 73

Leveraging the Power of Existing Systems 81

The Role of a Data Scientist 81

Feedback Loops 82

Making Data Accessible 84

Data Governance 85

Are You Data Ready? 89

Pitfalls 90

Action Checklist 94

Notes 94

Chapter 5 Prototyping 97

Is There an Existing Solution? 97

Employing vs. Contracting Talent 99

Scrum Overview 101

User Story Prioritization 103

The Development Feedback Loop 105

Designing the Prototype 106

Technology Selection 107

Cloud APIs and Microservices 110

Internal APIs 112

Pitfalls 112

Action Checklist 114

Notes 114

Chapter 6 Production 117

Reusing the Prototype vs. Starting from a Clean Slate 117

Continuous Integration 119

Automated Testing 124

Ensuring a Robust AI System 128

Human Intervention in AI Systems 129

Ensure Prototype Technology Scales 131

Cloud Deployment Paradigms 133

Cloud API’s SLA 135

Continuing the Feedback Loop 135

Pitfalls 135

Action Checklist 137

Notes 137

Chapter 7 Thriving with an AI Lifecycle 139

Incorporate User Feedback 140

AI Systems Learn 142

New Technology 144

Quantifying Model Performance 145

Updating and Reviewing the Idea Bank 147

Knowledge Base 148

Building a Model Library 150

Contributing to Open Source 155

Data Improvements 157

With Great Power Comes Responsibility 158

Pitfalls 159

Action Checklist 161

Notes 161

Chapter 8 Conclusion 163

The Intelligent Business Model 164

The Recap 164

So What are You Waiting For? 168

Appendix A AI Experts 169

AI Experts 169

Chris Ackerson 169

Jeff Bradford 173

Nathan S. Robinson 175

Evelyn Duesterwald 177

Jill Nephew 179

Rahul Akolkar 183

Steven Flores 187

Appendix B Roadmap Action Checklists 191

Step 1: Ideation 191

Step 2: Defining the Project 191

Step 3: Data Curation and Governance 192

Step 4: Prototyping 192

Step 5: Production 193

Thriving with an AI Lifecycle 193

Appendix C Pitfalls to Avoid 195

Step 1: Ideation 195

Step 2: Defining the Project 196

Step 3: Data Curation and Governance 199

Step 4: Prototyping 203

Step 5: Production 204

Thriving with an AI Lifecycle 206

Index 209

JEFFREY L. COVEYDUC is Vice President and Master Inventor at IBM. His diverse background consists of positions that encompass the creation of innovative, technologically advanced global AI solutions and client adoption.

JASON L. ANDERSON is a Partner and CTO with the data consultancy, Comp Three, where he established a new AI line of business. He is also a former IBM Cognitive Architect and Master Inventor. He received both BS and MS degrees in Computer Science from California Polytechnic State University, SLO.

We have reached a critical mass in the development of artificial intelligence. Thanks to products and services offered by the cloud, AI is now accessible even to smaller organizations or those with smaller budgets. And consumers are comfortable interacting with AI on a daily basis—think Apple's Siri, Netflix recommendations, and realtime GPS routing. With these two shifts, we see an elimination of the barriers to entry that once prevented many organizations from getting started with AI. Today, businesses know that AI is within their reach, and they know that their competitors, or disruptive startups, are working to leverage this new technology. AI is no longer an optional proposition.

We all need to think about implementing AI to stay competitive, but where do we start? Until now, there was no proven, step-by-step process to help businesses begin cutting costs and innovating using AI technology. In Artificial Intelligence for Business, Jeffrey L. Coveyduc and Jason L. Anderson provide just such a roadmap. This much-needed guide walks readers through the process of adopting AI technology, starting with identifying the opportunities most suited to AI solutions and leading all the way through deploying AI and iterating AI models for continuous improvement.

AI is inherently interdisciplinary, and, accordingly, this book takes an interdisciplinary approach. From a business perspective, leaders must understand that their most valuable resource is data. Locating (or, if necessary, creating), managing, and leveraging data resources is the name of the AI game. From a software development perspective, AI programming is very different from traditional application coding. If organizations and dev teams fail to understand the unique requirements of AI, their chances for success decrease. Readers will gain insight into each facet of AI and learn how to make them all work together for tangible value and innovation.

A PROVEN PROCESS FOR TRANSFORMING YOUR ORGANIZATION WITH AI TECHNOLOGY

The AI adoption journey is long, but the potential rewards are great. Many leaders have the drive and enthusiasm needed to get started with AI but no clear picture of how the process will unfold. Artificial Intelligence for Business minimizes the risk involved in making the transition to AI, both by providing concrete action steps and by identifying the most common pitfalls and how to avoid them. Such guidance could be the key to ensuring a profitable foray into the world of AI. Inside, you'll learn how to:

• Identify opportunities to reduce costs and capture market share using AI
• Locate the data you need to train AI models, and manage data assets professionally
• Create a functional AI prototype to limit risk and demonstrate the AI value proposition
• Confidently deploy and iterate your AI solutions in production
• Establish AI maturity using model libraries to capture profits and improve over time

This book is perfect for business leaders who want a high-level roadmap showing the way to proven success in the world of AI.