AI Risk Management Resource Library

A curated resource library with Industry and Actuartech resource focused on managing AI risk in insurance. Included for free whilst enrolled to A Practical Guide to AI Risk Management.

Interested in this course? Email us at [email protected]

Course curriculum

1. Welcome
1. Reference List
2. Summary of AI Risk Management Resources
1. Section Overview
1. 1.1.1 The Fairness of Machine Learning in Insurance: New Rags for an Old Man?
2. 1.1.2 Semi-supervised learning in insurance : fairness and active learning
3. 1.1.3 Anti-Discrimination Insurance Pricing: Regulations, Fairness Criteria, and Models
4. 1.1.4 Discrimination-Free Insurance Pricing
5. 1.1.5 Designing Fairly Fair Classifiers Via Economic Fairness Notions
6. 1.1.6 Paradoxes in Fair Machine Learning
7. 1.1.7 A Multi-Task Network Approach for Calculating Discrimination-Free Insurance Prices
8. 1.1.8 Insurers and regulators must stamp out discrimination in insurance pricing to ensure fairness for consumers
9. 1.1.9 Limits and concepts of the indirect discrimination
10. 1.1.10 IFoA: The hidden risks of being poor
11. 1.1.11 Fairness and Calibration in predictive models
12. 1.1.12 AI and Ethics in Insurance: A New Solution to Mitigate Proxy Discrimination in Risk Modelling (EAA e-Conference 2023)
13. 1.1.13 What is Fair? Proxy Discrimination vs. Demographic Disparities in Insurance Pricing
14. 1.1.14 Reducing bias in AI-based financial services
15. 1.1.15 How insurers can mitigate the discrimination risks posed by AI
16. 1.1.16 Fairlearn: Open source community driven project to help data scientist with the fariness of AI sysytems
17. 1.1.17 Model Fairness Assessment: The Fairness dashboard
18. 1.1.18 Bias in AI: What it is, Types, Examples & 6 Ways to Fix it in 2022
19. 1.1.19 Equality of Opportunity in Supervised Learning
20. 1.1.20 Proxy Discrimination in the Age of Artificial Intelligence and Big Data
21. 1.1.21 The Discriminating (Pricing) Actuary
1. 1.2.1 A Guide for Ethical Data Science
2. 1.2.2 Ethical Use of Artificial Intelligence for Actuaries - SOA
3. 1.2.3 IFoA APS X1: Applying Standards to Actuarial Work
4. 1.2.4 Exploring Ethics & AI within Financial Services Webinar
5. 1.2.5 Royal Statistical Society Paper - Additional information and resources: ethical data science
6. 1.2.6 AI Ethics - DataRobot
7. 1.2.7 Principles of AI Ethics (SAS GI Seminar 2019)
8. 1.2.8 IEEE - Ethics in Action in Autonomous and Intelligent Systems
9. 1.2.9 IEEE - Ethically Aligned Design
10. 1.2.10 AI and ethics in insurance: a new solution to mitigate proxy discrimination in risk modeling
11. 1.2.11 Data & Ethics: How to establish a sustainable, data-driven business model in life insurance
12. 1.2.12 Responsible Use of Data in the Digital Age: Customer expectations and insurer responses
13. 1.2.13 ProActuary: Culture, Professionalism and Digital Actuary
14. 1.2.14 AI risk and ethics forum series: Trustworthy AI
15. 1.2.15 IFoA APS X2: Review of Actuarial Work
16. 1.2.16 NZ Police - Safe and ethical use of algorithms
17. 1.2.17 AI Ethics and Insurance (SOA)
1. 1.3.1 Mind the Gap - Safely Incorporating Deep Learning Models into the Actuarial Toolkit by Ronald Richman
2. 1.3.2 Big data and data profiling in the insurance industry - RPC
3. 1.3.3 Believing the bot – model risk in the era of deep learning
4. 1.3.4 Machine Learning: A Practical Guide To Managing Risk
1. 1.4.1 Closing the gap: How data and technology can help reduce health inequalities in London?
2. 1.4.2 The Alan Turing Institute: Understanding artificial intelligence ethics and safety
3. 1.4.3 ELISE - European Network of AI Excellence Centres
4. 1.4.4 The impact of artificial intelligence on the future of workforces in the European Union and the United States of America
5. 1.4.5 The rise of AI in insurance is it a double edged sword
6. 1.4.6 Enable Trustworthy AI (British Standards Institute)
7. 1.4.7 Decrypting AI: Opportunities of AI in insurance
8. 1.4.8 Public Lecture - Deep Learning: What Could Go Wrong? by Des Higham - 25/04/23
9. 1.4.9 Man vs machine - The Actuary Magazine
10. 1.4.10 Artificial Intelligence and Algorithmic Liability
1. 1.5.1 Explanation in Artificial Intelligence: Insights from the Social Sciences
2. 1.5.2 Interpretable Machine Learning: A Guide for Making Black Box Models Explainable - Christoph Molnar
3. 1.5.3 Four Principles of Explainable Artificial Intelligence
4. 1.5.4 Explaining predictive models using Shapley values and non-parametric vine copulas
5. 1.5.5 Explaining individual predictions when features are dependent: More accurate approximations to Shapley values
6. 1.5.6 AI Standards Hub - Towards transparent and explainable AI
7. 1.5.7 XEMP Prediction Explanations - DataRobot
8. 1.5.8 Synthetic Dataset Generation of Driver Telematics
9. 1.5.9 DARPA’s Explainable Artificial Intelligence (XAI) Program
10. 1.5.10 A Comprehensive Taxonomy for Explainable Artificial Intelligence: A Systematic Survey of Surveys on Methods and Concepts
11. 1.5.11 Explainable Artificial Intelligence (XAI) in Insurance
12. 1.5.12 Understanding algorithmic decision-making: Opportunities and challenges
13. 1.5.13 Stop Explaining Black Box Machine Learning Models for High Stakes Decisions and Use Interpretable Models Instead
14. 1.5.14 Interpretable Machine Learning for Insurance: An Introduction with Examples
15. 1.5.15 Why Should I Trust You? - Explaining the Predictions of Any Classifier
16. 1.5.16 Explaining decisions made with AI
17. 1.5.17 Enforcing Interpretability and its Statistical Impacts: Trade-offs between Accuracy and Interpretability
18. 1.5.18 A Unified Approach to Interpreting Model Predictions
19. 1.5.19 Explainable Artificial Intelligence: a Systematic Review
20. 1.5.20 The Mythos of Model Interpretability
1. 1.6.1 'Machine Decisions': Governance of AI and Big Data Analytics
2. 1.6.2 From Roadmap to Reality: Insights from Industry on Advancing AI Assurance
3. 1.6.3 Industry temperature check: barriers and enablers to AI assurance
4. 1.6.4 Cert AI: Trustworthy AI and Compliance
5. 1.6.5 Trusted AI 101: A Guide to Building Trustworthy and Ethical AI Systems
6. 1.6.6 AI transparency in financial services - why, what, who and when?
7. 1.6.7 Model Artificial Intelligence Governance Framework
8. 1.6.8 Auditing the AI Auditors: A Framework for Evaluating Fairness and Bias in High Stakes AI Predictive Models
9. 1.6.9 Governance of Artificial Intelligence in Finance
10. 1.6.10 AI Governance Principles: Towards ethical and trustworthy AI in the European Insurance sector
11. 1.6.11 Guidance on the AI auditing framework: Draft guidance for consultation
12. 1.6.12 Non-regulatory Approaches to AI Governance: Spotlight on AI Assurance Techniques and Standards (EAA e-Conference 2023)
13. 1.6.13 Human Rights, Democracy, and the Rule of Law Assurance Framework for AI Systems: A Proposal
14. 1.6.14 How ISO/IEC 42001 guides organisations toward Trustworthy AI
15. 1.6.15 Using AI in the public sector: New comprehensive guidance
16. 1.6.16 KPMG: Trust in algorithms with AI in control
17. 1.6.17AI Standards Hub: Exploring safe, secure, resilient, and trustworthy AI
18. 1.6.18 AI in financial services: Are you meeting the regulators’ expectations?
19. 1.6.19 Auditing algorithms: the existing landscape, role of regulators and future outlook
20. 1.6.20 German Standardization Roadmap on Artificial Intelligence
21. 1.6.21 A European Strategy for Artificial Intelligence
22. 1.6.22 Restoring trust in audit and corporate governance
23. 1.6.23 Casualty Actuarial and Statistical (CAS) Task Force Regulatory Review of Predictive Models : White Paper
24. 1.6.24 2020 Algorithm Charter for Aotearoa New Zealand
25. 1.6.25 Algorithm Audit European Knowledge Platform
26. 1.6.26 Wired: Want to Prove Your Business Is Fair? Audit Your Algorithm
27. 1.6.27 A toolbox to better understand your ML results – With application in insurance pricing
28. 1.6.28 The Aletheia Framework
29. 1.6.29 Deloitte: Algorithm Insights - Risk and control considerations
30. 1.6.30 Assessing model risk in practice (The Actuary Magazine)
31. 1.6.31 The how and why of data science risk management (EAA Conference 2022)
32. 1.6.32 Supervisory guidance on model risk management
33. 1.6.33 Risk and Control: Ideas for a better tomorrow
34. 1.6.IFoA: Model Risk - Daring to open up the black box
35. 1.6.35 Algorithm assessment report
36. 1.6.36 What is AI Governance?
37. 1.6.37 Advancing accountability in AI
1. Section Overview
1. 2.1.1 IFoA: Ethical and professional guidance on Data Science
2. 2.1.2 Recommendation of the Council on Artificial Intelligence (OECD)
3. 2.1.3 IFoA: The Actuaries' Code
4. 2.1.4 IFoA: Guidance on the application of Technical Actuarial Standard 100
5. 2.1.5 AAE: European Standards of Actuarial Practice 1
6. 2.1.6 AAE: European Standards of Actuarial Practice 2
7. 2.1.7 FRC: Technical Actuarial Standard 100: Principles for Technical Actuarial Work
8. 2.1.8 FRC: Technical Actuarial Standard 200: Insurance
9. 2.1.9 FCA: FG22/5 Final non-Handbook Guidance for firms on the Consumer Duty
10. 2.1.10 Queen's Speech 2022: data protection implications
11. 2.1.11 Queen's Speech 2022 - DMA calls for urgent publication of data reforms by UK Government
12. 2.1.12 IFoA: New actuarial regulation proposals welcome - but the devil is in the detail
13. 2.1.13 Equality Act 2010
14. 2.1.14 FCA: Treating Costumers Fairly - Principles of good regulation
15. 2.1.15 Regulatory Technologies to Enhance Conduct
16. 2.1.16 AAE: European Actuarial Note 1 on ESAP 3 and ORSA
17. 2.1.17 AAE: European Actuarial Note 2 on Actuarial Function under IORP II
18. 2.1.18 ASSA: COVID-19 Considerations for Life Assurance Actuaries
19. 2.1.19 ASSA: APN 401 - Establishing Technical Provisions for Short-Term Insurers
20. 2.1.20 ASSA: APN 901: General Actuarial Practice
1. 2.2.1 Guidance on AI and data protection
2. 2.2.2 EU AI Act
3. 2.2.3 DP5/22 - Artificial Intelligence and Machine Learning
4. 2.2.4 Algorithmic Transparency Recording Standard (ATRS)
5. 2.2.5 EU Artificial Intelligence Act: The European Approach to AI
6. 2.2.6 OECD: The AI Wonk
7. 2.2.7 POPIA and the insurance industry (RSA)
8. 2.2.8 General Data Protection Regulation (GDPR)
9. 2.2.9 AI Standard Hub: Introducing the Hub’s work programme on Trustworthy AI
10. 2.2.10 A pro-innovation approach to AI regulation
11. 2.2.11 UK Regulators' Path for AI Starts With Auditing Algorithms
12. 2.2.12 BBC: UK rules out new AI regulator
13. 2.2.13 Big Data and Insurance: Implications for Innovation, Competition and Privacy
14. 2.2.14 Big Data in the Insurance Industry: Leeway and Limits for Individualising Insurance Contracts
15. 2.2.15 Big Data And Algorithms In Actuarial Modelling And Consumer Impacts
16. 2.2.16 Big data in insurance
17. 2.2.17 Data: A new direction
18. 2.2.18 Artificial Intelligence in financial services: An evolving regulatory focus
19. 2.2.19 Technical methods for regulatory inspection of algorithmic systems
20. 2.2.20 Big data meets artificial intelligence: Challenges and implications for the supervision and regulation of financial services
1. Section Overview
1. 3.1.1 SOA: Actuarial Technology, A Roundtable Discussion on Current Issues
2. 3.1.2 IFoA: Actuarial Software and Calculations - Professional Responsibilities
3. 3.1.3 What is AI Auditing?
4. 3.1.4 Global AI Hub
5. 3.1.5 AI and The Opportunities And Challenges it presents to Insurability - AAE
6. 3.1.6 Deloitte: Risk Technology Strategy, Selection and Implementation
7. 3.1.7 Open Risk Management, Open Source - The Tools of the Trade
8. 3.1.8 The Importance of Centralization of Actuarial Modeling Functions, Part 2: DevOps - The Path to Actuarial Modernization and Consolidation (SOA)
9. 3.1.9 Hacking Insurance AI algorithms with the French regulator
1. Section Overview
1. 4.1.1 Deloitte: Managing algorithmic risks - Safeguarding the use of complex algorithms and machine learning
2. 4.1.2 JFAR Risk Perspective Report: 2021
3. 4.1.3 JFAR Risk Perspective Report: 2022
4. 4.1.4 IFoA Data Science Research Station
5. 4.1.5 Data Science in Insurance: Opportunities and Risks for Consumers
6. 4.1.6 ORCAA: Its the age of Algorithm and we have arrived unprepared
7. 4.1.7 Pew Research Center: Auto insurance discounts and monitoring
8. 4.1.8 Insurance market failing vulnerable consumers: IFoA report
9. 4.1.9 Statistical Foundations of Actuarial Learning and its Applications
10. 4.1.10 Are there monsters in our models?
11. 4.1.11 The Use of Artificial Intelligence and machine Learning in UK Actuarial Work
12. 4.1.12 Thematic Review Report: Actuaries using data science and artificial intelligence techniques
13. 4.1.13 From Bias to Black Boxes: Understanding and managing the risks of AI - an actuarial perspective
1. Section Overview
1. 5.1.1 Do Not Answer: A Dataset for Evaluating Safeguards in LLMs - Paper
2. 5.1.2 Do Not Answer: A Dataset for Evaluating Safeguards in LLMs - GitHub Repository
3. 5.1.3 Baker McKensie: China: New interim measures to regulate generative AI
1. 5.2.1 ActuaryGPT: Applications of Large Language Models to Insurance and Actuarial Work
2. 5.2.2 A responsible approach to using AI in corporate reporting Guidance for Boards and management on approach and disclosure

About this course

Free
184 lessons
1.5 hours of video content

AI Risk Management Resource Library

Course curriculum

Overview

Contents

1. Practical AI Risk Management

1.1 Fairness

1.2 Ethics and Professionalism

1.3 Machine Learning (ML) in Actuarial Work

1.4 Artificial Intelligence (AI)

1.5 Interpretability & Explainability

1.6 Governance

2. Regulation

2.1 Conduct

2.2 AI Governance and Data Protection

3. Industry Best Practice

3.1 Technology

4. Current AI Landscape: An Actuarial Perspective

4.1 Wider Industry Themes

5. LLMs

5.1 LLM Risks and Risk Management

5.2 LLMs for Actuarial Work

About this course

Interested in this course?