Learn how to manage risks associated with artificial intelligence (AI) in an actuarial context, with a focus on machine learning.
If you're looking for a practical guide to performing AI risk management, this course is ideal for you. The course offers a discussion of different guidelines for AI risk management, walking through explainability techniques, and recommending best practice. An individual subscription gives you 3 months’ online access to:
As Well As
Our Industry, Actuartech, and Risk Management Resource Libraries which feature curated additional content to assist you on your data science journey.
With the use of examples, we introduce delegates to the new types of risks resulting from AI, with a focus on technical considerations, including interpretability, explainability, validation, and transparency. We provide examples to make sense of apparent ‘black box’ models; and ways to validate certain types of machine learning models.
We will present participants with pre-run models; based on publicly available data; and provide hands-on examples to interpret these models. Note that familiarity with Python, Jupyter Notebooks and the ability to code foundational Python is recommended as delegates will be guided through advanced techniques.
We will also explore industry examples of ethics, professionalism and regulatory principles within the context of AI, and provide delegates with curated resources and summarised regulatory requirements.
The main challenges actuaries are facing will be discussed, particularly as it relates to the use of AI for modelling and analytics. These themes include:
Further themes such as the individualisation of risk assessment and the lack of relevant skills are also discussed in detail in the Industry Paper available in the course. The course also concludes with a discussion on the use of Large Language Models (LLMs) for actuarial work.
In Chapter 1 we introduce the course and define AI within the context of actuarial work.
Chapter 2 discusses key concerns and considerations for AI risk management, and looks at practices to help manage the risks in machine learning investigations, including transparency, validation techniques, and explainability. This chapter also discusses the governance of AI.
Chapter 3 outlines the regulatory journey of the use of AI by actuaries, focused on the UK, and considers global regulatory requirements for AI systems.
In Chapter 4 we summarise best practice for the use of AI in insurance and provide resources on the use of AI, with Chapter 5 providing an actuarial perspective on the current AI risk landscape.
Chapter 6 discusses some considerations regarding the use of large language models for actuarial work.
The Appendix contains the downloadable notebooks and an optional assignment, along with key course references.
This course is suitable for anyone with some familiarity with R and Python who would like to explore how AI risk can be managed within an actuarial context, including:
AI is changing the actuarial operating model, as it holds the opportunity to improve current processes by implementing AI, particularly machine learning, techniques. There are a lot of benefits as implementation and good practices of these techniques can ensure reproducibility, code control, out of the box thinking, and sharing development in a transparent way. But this could only be achieved if it is delivered in a risk-controlled manner; risk management and risk management considerations has to keep up with the change in operations.
We have tailored packages available to ensure that corporate teams have the option to attend structured live lessons by our tutors, and the option to request a practical assignment and bespoke feedback. Invoicing option available.
