Responding to Treasury's review of AI and Australian Consumer Law, the Institute has drawn on recommendations made in prior extensive submissions to Government on AI including the need for regulation to be outcomes focused, prioritisation of existing regulation and guidance rather than new regulation and the creation of a centralised expert body to consider AI governance, regulation and guidance.

The Institute acknowledges the complexities of regulating AI systems, welcoming the acceptance of the OECD definition of AI systems and more detailed outline of risk dimensions for AI systems. We recommend further refinement of the proposed guardrails including more detail in the risk classification system and an understanding of the costs associated if the guardrails are implemented.

The Institute's response encourages Government to prioritise clarity of existing principles-based regulation across a range of sectors before creating new-technology focused regulation specifically for the use of AI. Existing regulation should be reviewed in the context of AI and guidance provided as a means of reducing regulatory complexity.

Artificial Intelligence is having a significant impact on the insurance industry, with rapid development and investment across the value chain. While AI presents opportunities for efficiency and improved service, it also brings new risks and regulatory challenges. We will explore three perspectives on the emerging relation between AI and insurance, with a focus on how regulatory and legislative changes are likely to impact insurers’ own use of AI, change the risk landscape for existing lines and introduce opportunities for new product innovations. We’ll cover: Impacts of emerging regulations and standards on insurance applications: We'll explore the rapidly expanding legislative, regulatory and standards environment. Examples include specific AI legislation in the EU and US, emerging regulations in Australia and updates to existing legislation to explicitly capture algorithmic and AI risks, such as the proposed changes to the Privacy Act 1988 (CMW). Complementing the regulatory development are standards and guidelines on appropriate application and governance of AI, including the ISO 42 group of standards that seek to provide a rigorous framework for system validation and audit, akin to existing standards around Cyber risk. AI-induced risks on existing insurance lines: Here we’ll address the ripple effects of AI on established lines. Examples include the potential for breach of duty claims impacting Directors and Officers insurance, complex changes to liability risk for organisations that include AI in their products and increased risks of privacy and data breaches impacting risk exposures for cyber cover. We'll discuss relevant international developments and examples, including the EU's ambitious Artificial Intelligence Liability Directive, which seeks to update the EU liability framework to make it easier for individuals to bring claims for harms caused by AI. Insurance solutions for emerging AI risks: In this section, we'll cover the potential for insurance products that cover new and emerging AI risks, including existing emerging examples. This includes nascent product-warranty like cover, and cover for consequential losses arising from underperformance of AI models. We'll discuss potential for cover against insurability criteria, and some of the associated likely product, underwriting and claims challenges.

The adoption of artificial intelligence (AI) across industries has led to the widespread use of complex black-box models and interpretation tools. This paper proposes an adversarial framework to uncover the vulnerability of permutation-based interpretation methods for machine learning tasks, with a particular focus on partial dependence (PD) plots. This adversarial framework modifies the original black box model to manipulate model predictions for instances in the extrapolation domain, resulting in PD plots that can hide discriminatory behaviors while maintaining the prediction accuracy of the original model. This framework can produce multiple fooled PD plots via a single model. By using real-world datasets including an auto insurance claims dataset and COMPAS dataset, our results show that it is possible to intentionally hide the discriminatory behaviour of a predictor and make the black-box model appear neutral through interpretation tools like PD plots while retaining almost all the predictions of the original blackbox model.

Traditional aggregate reserving techniques assume that each data triangle used is a homogeneous group of claims. However, the practical constraints of management and reporting requirements often hinder the adoption of a statistically optimal approach to segmentation. There is often a reliance on legacy class hierarchies, business processes or expert judgement due to the difficulty of determining an optimal segmentation. This can reduce homogeneity in triangles and affect the performance of aggregate reserving techniques. To address these challenges, we propose a framework and methodology for automated segmentation to allocate individual claims in a reserving class to homogeneous subgroupings more suitable for triangle-based projections. The proposed approach involves use of machine learning techniques to: • Evaluate claim features that can relate to claim development in a statistically significant way. • Implement an algorithm to segment claims. • Assess and compare different segmentations on their impact on claim reserve accuracy. We also explore the practicalities for constructing models in a way that allows managing segmentation changes across reporting periods. Overall, we propose a methodology and practical framework for leveraging clustering techniques as part of a reserving process to improve accuracy and reduce the need for intervention of actuaries managing insurers’ claims reserves.

In an era where financial literacy and risk management are paramount, understanding the motivations behind life and health insurance purchases is crucial for the industry. This report explores the behavioral factors influencing insurance purchase among Gen Z and Millennials for forward-looking purposes, leveraging a rich dataset spanning 22 markets across six continents. Employing both traditional economic regression and state-of-the-art machine learning models, this study offers a dual perspective on market behaviors and purchasing trends. Our comprehensive study is based on data from the Global Consumer Study 2023-24, a global survey of over 12,000 individuals carried out by SCOR’s Digital Solutions division. It delves into demographics, insurance preferences, personal circumstances, and digital engagement. With an analytical focus on younger generations, the findings illuminate key predictors for insurance purchase: 1. Insurance Knowledge: A foundational factor, where higher understanding correlates with increased insurance purchases. 2. Marital Status: Highlighting a trend where married individuals are more prone to purchasing life and health insurance than their single peers. 3. Employment: Demonstrating that those with full-time and part-time employment are more inclined to invest in life and health insurance. 4. Property Ownership: Linking property ownership with a higher likelihood of having insurance coverage. 5. Urban Residency: Observing that urban dwellers are more apt to purchase insurance. 6. Age Dynamics: For Gen Z and Millennials, relatively young segments of the general population, there’s a trend where interest in life insurance slightly increases with age, while the inclination to purchase health insurance experiences a marginal decline. 7. Sex Preferences: Men show a predilection for life insurance, whereas women display a preference for health insurance. The study delves into the inclination towards recent insurance purchases, particularly noting an uptick among younger generations endowed with high insurance literacy. It has been observed that individuals who have made claims are also more inclined to engage in recent purchases. This trend underscores the generations’ readiness to adopt online solutions, especially those prioritizing cost efficiency and effectiveness. Among middle-aged groups, a notable interest in utilizing health apps correlates with a higher likelihood of recent insurance acquisitions. The findings highlight a major move towards using digital methods to buy insurance, especially noticeable in major urban cities. This move to online platforms is a big change in the insurance world, requiring companies to skillfully adapt to these changes to remain relevant. This is especially important as they try to attract the growing buying power of younger generations. In conclusion, the future trajectory of the insurance market hinges on a nuanced understanding of these evolving trends. By embracing technology-driven solutions and catering to the preferences of an informed, tech-savvy generation, insurance companies can establish a strong presence in the constantly changing market.

The Institute recommends prioritising the clarity of existing regulation where artificial intelligence decisions reflect equivalent human decisions. We recommend a risk-based approach to regulation to balance potential gains made using artificial intelligence against the risks associated with its use

Responding the Government’s Report on the Privacy Act Review, the Institute provides feedback on four areas 1.Expanding the scope of the Privacy Act to include inferred or generated information; 2. Rights of Portability; 3. Automated Decision-Making; and Explanatory Materials to guide institutions and practitioners of their expected conduct.

This Guidance Resource has been developed by the Australian Human Rights Commission (the Commission), in partnership with the Actuaries Institute (Institute), to provide guidance to professionals and businesses on complying with federal anti-discrimination legislation in relation to use of artificial intelligence (AI) in insurance pricing and underwriting decisions.

Questions arising from the recent competition compliance training sessions are set out in this document.

This Green Paper explores how big data is transforming the insurance industry and the implications for the cost and availability of insurance

This paper from the COVID-19 Mortality working Group looks at a number of ways COVID-19 mortality has been measured around the world and in Australia by examining various datasets and discussing different statistical techniques observed. We then take a closer look at the mortality experience of Australia during 2020 and 2021, including experience by Cause of Deaths (including COVID-19), by age group, and experience during different COVID-19 waves . And finally, we take a brief look at the impact of COVID-19 on long term illness, based on studies from around the world.

On the issue of insurance discrimination, a grey area in regulation has resulted from the growing use of big data analytics by insurance companies – direct discrimination is prohibited, but indirect discrimination using proxies or more complex and opaque algorithms can be tolerated without restrictions. This phenomenon has recently attracted the attention of insurance regulators all over the world, and stricter insurance discrimination regulations are being discussed and considered by regulators. Meanwhile, various fairness criteria have been proposed and flourish in the machine learning literature with the rapid growth of artificial intelligence (AI) in the past decade, which mostly focus on classification decisions. In this paper, we introduce the fairness criteria that are potentially applicable to insurance pricing as a regression problem to the actuarial field, match them with different levels of potential and existing antidiscrimination regulations, and implement them into a series of existing and newly proposed anti-discrimination insurance pricing models, using both generalized linear models (GLMs) and Extreme Gradient Boosting (XGBoost). Our empirical analysis compares the outcome of different models via fairness-accuracy trade-off and shows the impact on customer behavior and solidarity.

Exchange of information is a critical part of insurance pricing and underwriting. Traditionally, this is in the form of mandatory question sets, which the prospective insured person must answer to a suitable level of reliability before obtaining a quote for cover. In Australia, the Insurance Contracts Act sets out some rules around this, and other analogous systems exist in various other countries around the globe. The traditional manner of data collection had inherent practical limits. Questions had to be easily understood by laypeople, readily answerable by them, and not so extensive as to be off-putting. With the advent of open data regimes around the globe, many of these traditional limitations may be reduced or removed. By a mere press of a button, consumers may be able to share extensive and unprecedented data with an insurer, in order to automatically and accurately answer detailed questions that they might not necessarily understand or be able to answer if asked directly. In this paper, we analyse whether open data regimes can be used in this manner to replace existing underwriting questions or to create new ones. We then examine the impact that this change may have on various cohorts of customers, particularly considering the potential impact on those without access to data, who may be more likely than average to be otherwise vulnerable or disadvantaged. We suggest thematic areas to consider for further guidance or reform, based on our analysis.

Predicting customer churn is an important consideration for any business, including financial service businesses, because costs of acquiring new customers far outweigh costs of retaining existing ones. Our daily interactions with Siri, Alexa, Hey-Google, and Bixby, which are Natural Language Processing (NLP) based automation systems are currently treated as just another cool feature in our everyday lives. Imagine using this cool feature to solve a fundamental problem for a business – preventing customer churn. Different customers exhibit different behaviours and preferences and cancel their subscriptions for a variety of reasons. Most existing models predict customer churn by using demographic and transactional data of customers, which may not contain a full reflection of customers’ intentions. In this paper, a customer churn prediction model is developed using NLP by extracting features and patterns in unstructured data available against customer policies. These unstructured datasets are typically text, calls, and notes, and thanks to the advancement of NLP technology, these datasets can now transform into key information from which we can infer intention for churn. Existing commercial NLP models which predict customer intention based on text, are still in their infancy and researchers are still investigating how to improve such models. With fast emergence of new NLP features, many have become outdated. A case study is presented in this paper to predict customer churn and reason for intention to churn using call data which may not be possible using structured data alone. The model proposed in this paper uses recently available NLP tools and features to develop a customer churn prediction model. The model uses keyword matching to mine expressions of interest and profiles of people corresponding to customer criteria. The proposed model takes advantage of available pre-trained NLP models to perform sentiment analysis. A set of reference sentiments are manually generated and compared with the customer conversation to find the similarity as an index. This index is used as a threshold for a classifier model to identify the reasons for churn for any conversation. The performance shows that NLP has the potential to provide a detailed understanding customers’ churn behaviour including why a customer chose to churn.

Retail insurance pricing has been a prolonged and complex process involving many technical and practical considerations. The rapid market changes require insurers to not only have an established framework to conduct pricing reviews, but also the capabilities to translate data into market pricing responses in an accurate and efficient manner. One particular challenge faced by the insurers is how to deal with data from most recent period. The latest experience may include valuable insights into emerging market trends yet is often underdeveloped. Another challenge is the GBM modelling results are not directly implementable since the rating engine can only accommodate the GLM-like rating tables. In this paper, we propose an intelligent pricing approach to better utilizing machine learning models to improve insurers’ pricing capabilities, which could be well integrated into insurers’ existing pricing algorithms. The approach aims to overcome the previously highlighted two challenges and to enable an efficient risk modelling and pricing delivery process, by directly leveraging the machine learning modelling results. A case study is presented to demonstrate the viability and highlight the advantages of the intelligent pricing approach using actual insurance claim data. The accuracy and efficiency nature of the pricing solution is expected to largely boost insurers’ pricing capabilities under rapid changing market conditions.

Recent years have seen rapid increase in the application of machine learning to insurance loss reserving. These machine learning methods are hungry for data. While the ultimate objective of these methods is application to real data, the availability of synthetic data is important for at least two reasons: (i) real data sets, especially of granular nature and of large size, are in short supply in the actuarial literature for reasons of confidentiality, (ii) knowledge of the data generating process (impossible with real data) assists with the validation of the strengths and weaknesses of any new methodology. Against this background, we introduce a simulator of individual claim experience, called SPLICE (Synthetic Paid Loss and Incurred Cost Experience). On a high level, SPLICE consists of a paid loss unit (claim payments) and an incurred loss unit (case estimates). SPLICE simulates individual claims at the transactional level, i.e. key dates associated with a claim (e.g. notification and settlement dates), individual claim payments and revisions of case estimates. An individual claim’s transactions are generated in a manner that is intended to reflect transactional sequences observed in practice. Our simulator is publicly available, open-source (on CRAN), and fills a gap in the non-life actuarial toolkit. The simulator specifically allows for desirable (but optionally complicated) data features typically occurring in practice, such as superimposed inflation and various dependencies between claim features. For ease of use, SPLICE comes with a default version that is loosely calibrated against a specific real CTP portfolio and that has a structure suitable for most lines of businesses with some amendments. However, the modular structure of SPLICE ensures that the user has full control of the evolution of an individual claim (occurrence, notification, timing and magnitude of individual partial payments and revisions for case estimates, closure). Indeed, thanks to this flexibility, SPLICE may be used to generate a collection of data sets that provide a spectrum of complexity. Such a collection may be used to present a model under test with a steadily increasing challenge.