Executive summary i

Figures vi

Tables vi

Boxes vi

Abbreviations vii

Acknowledgements viii

Chapter 1. Introduction 1

  1.1. Scope 3

Chapter 2. AI use cases in supply chain operations 5

  2.1. AI in transportation and logistics 6

  2.2. AI in warehousing and inventory management 10

  2.3. Analysis of AI incidents and lawsuits 12

Chapter 3. Insights from stakeholder interviews 15

  3.1. AI adoption remains uneven, but current use is concentrated in optimisation, forecasting, monitoring and document processing 15

  3.2. AI adoption is being driven by efficiency and visibility, but constrained by data, capability and trust 16

  3.3. AI is seen as a source of new vulnerabilities, with continued reliance on human judgement and intervention 17

  3.4. The insurance market is beginning to adapt to AI, but there is no settled approach yet 18

  3.5. Responsibility for AI-related failures remains unclear, and governance is still fragmented 19

Chapter 4. Five-year AI adoption outlook 21

  4.1. Accelerating AI adoption 22

  4.2. AI adoption constraints 25

  4.3. Likely adoption trajectory across deployment levels 27

Chapter 5. AI adoption implications for insurance 29

  5.1. Potential losses from AI risk 29

  5.2. Exclusions for AI risks 30

  5.3. Affirmative cover for AI risk 32

  5.4. Silent cover for AI risk and implications for policyholders 33

  5.5. Insurance lines most likely to respond to AI-related supply chain losses 33

Chapter 6. Aggregation scenarios 35

  6.1. Aggregation risk as the core insurance problem 35

  6.2. Scenarios 38

Chapter 7. Policy recommendations 51

  7.1. Supply chain operators and AI & technology providers 51

  7.2. Insurers 54

  7.3. Regulators 55

References 58

Annex A. Research methods 65

  A.1. Literature review 65

  A.2. Characterising risk exposure 68

  A.3. Methodology for stakeholder interviews 71

  A.4. Scenario development 72