All Actuaries Summit 2026 Program

Explore the tools required to enable leaders to respond to complex challenges in health and general insurance. The sustainability challenge is no longer just financial, it is structural, behavioural and cultural. Suhit Anantula, AI Strategy and Human-Centred Innovation expert, and Jacki Johnson, Non-Executive Director, Community First Bank, explore how leaders can shift between planning, experimentation and sensing to build more resilient, customer-aligned insurance models.

Multi-rhythm leadership is the capability to move deliberately between structured planning, iterative learning and real-time sensing.

Leave with practical tools to lead with confidence amid rising costs, natural disasters, climate volatility and rapidly evolving customer expectations.

The rapid advancement of machine learning has provided an opportunity to transform the modeling techniques in actuarial analytics. Novel machine learning methods, especially deep learning, have demonstrated versatile modeling capability and superior predictive performance compared to traditional actuarial approaches such as Generalized Linear Models. However, the widespread adoption of deep learning techniques in the insurance industry is often hindered by the lack of model interpretability. The complex and opaque nature of these models makes it difficult to understand how predictions are generated, posing a significant challenge for pricing actuaries who must be able to explain and justify model outcomes for technical, commercial, and regulatory purposes.

Developing an inherently interpretable deep learning model for general insurance pricing is highly desirable, as it enables actuaries to harness the strong predictive power of deep learning while ensuring the model remains commercially viable and transparent. However, building such a model is challenging due to the absence of a universally accepted definition of interpretability. Establishing a rigorous, industry-wide framework to define and assess model interpretability is therefore essential. Moreover, beyond interpretability, various practical considerations (e.g. smoothness, monotonicity) are critical requirements for general insurance pricing models.

This research makes two primary contributions:

1. Interpretability framework for insurance pricing: We establish a rigorous framework that provides a concrete definition and mathematical formulation of interpretable pricing models. The framework articulates how interpretability can be formally expressed and outlines the interpretability and practical requirements that are essential in the insurance pricing context. This serves as a foundation for developing a common standard that can be adopted across the industry to define, design, and evaluate the interpretability of pricing models.

2. Actuarial Neural Additive Model (ANAM): We introduce ANAM, an inherently interpretable deep learning model for general insurance pricing. ANAM integrates the Neural Additive Model with several key enhancements within both its architecture and training process. These features ensure that the model satisfies actuarial interpretability standards while maintaining strong predictive performance and practical applicability.

Comparisons of predictive performance are conducted against both traditional actuarial models and state-of-the-art machine learning techniques using synthetic and real-world insurance datasets. The results demonstrate that the proposed model consistently outperforms alternative approaches in most scenarios while maintaining full transparency in its internal mechanisms, highlighting its strong interpretability and predictive power.

Concern over intergenerational equity has grown over time, with the three key drivers of inequity (income & wealth, housing, and the environment) facing continued attention. This presentation will combine findings from recent analysis by the authors as well as other key research pieces to provide a picture of intergenerational equity issues; this topic is highly relevant to proposals for economic reforms. Our presentation will cover:

 Key findings from the Australian Actuaries Intergenerational Equity Index (AAIEI) update to 2025
 Deeper analysis following the update into the housing domain and the lifecycle of government funding flows by age
 Other key research into intergenerational inequity including discussion of the implications and proposed solutions.

The AAIEI (https://tinyurl.com/4tfh6h83 and https://tinyurl.com/y3evt62m) tracks 24 indicators across six broad domains that relate to wealth and wellbeing (Economic and Fiscal, Housing, Health and disability, Social, Education and Environment). The absolute change as well as the relative change is tracked for three age groups. This provides a means to explore changes in intergenerational equity over time. We recently updated the index to 2025 and will highlight key findings.

Alongside the update we carried out analysis into the highly topical area of housing, which is marked by poor affordability, falling home ownership rates and longer durations renting. As one bank CEO recently expressed it “If [the housing market] doesn’t change and you live in Sydney, you’ll never know your grandkids”. By exploring data behind the indicators we will present insights on rental affordability plus home ownership projections. We will discuss findings and relevant policy implications.

A further analysis area was the ‘lifecycle’ view of how money (particularly government money) flows by age; younger age groups receive support (e.g. education), working-aged adults tend to be net taxpayers (taxes paid are greater than government services and benefits, on average), and older people net beneficiaries (e.g. through the age pension). Existing research paints a good picture of this lifecycle. We built on this by updating social spending estimates to capture recent programs (such as the NDIS rollout). We will discuss findings and policy implications for fiscal sustainability over the lifecycle.

The final component of our presentation will cover key research by other authors, the Intergenerational Report, and takeaways for actuaries in terms of policy proposals and reforms to watch for over coming years.

Are you looking to grow in new classes or geographies? Are you looking to start a new book of business? Or, do you simply want to understand broad trends better? It’s tough to build pricing and underwriting models when you lack data. This presentation will show you how you can use interesting external data sets to supplement your own data, with examples. Please come with the willingness to be creative, scrappy, and think outside-the-box.

Key Messages
 If you have no claims and exposure data, you can get creative in using publicly available data sources. Insurtechs have done this with great success, and you can too.
 We will share some tangible examples of creative use cases.
 How to supplement industry standard models with non-traditional data to develop your company’s own view of risk.

The increasing adoption of Electric Vehicles (EVs) and Personal Mobility Devices (PMDs) across Australia is leading to a diverse and everchanging landscape. The increased adoption has enabled sustainable and efficient alternatives to traditional transportation methods; promoting healthier lifestyles whilst creating efficient solutions to common infrastructure issues including traffic congestion and the ‘last mile’ problem. However, the rapid adoption of EVs has also introduced various new and distinct challenges for insurers, regulators and policymakers.

Whether it's food delivery, tourism or routine travel; the modernisation of the motor industry brought upon by the increased electric vehicle usage is creating a vastly different operating compared to what has been observed historically. Different road and vehicle laws, licensing, scheme and product design has created significant experience differences between Australian states. This has created unique operating environments in which insurers must navigate.

Personal Mobility Devices are small, electrically powered device and are generally designed for a singular user to travel short distances. They typically include devices such as electric scooters (e-scooters), electric bikes (e-bikes) and electric skateboards (e-skateboards). PMDs have seen increased usage over recent years which can partially be attributed to the ease of access caused by the entrance of shared mobility companies into the market. This has led to PMDs becoming heavily used, particularly in the food delivery industry where delivery drivers leverage them to efficiently navigate urban areas.

PMDs can exhibit vastly different risk characteristics to other forms of conventional transport due to their unconventional design, distinct rider behaviour and different operating environments. Road infrastructure and urban environments, state-based legislation, interactions with pedestrians and other road users and the implementation of telematics all have the propensity to further influence risk profiles creating further complexities for insurers.

With a focus on electric bikes and electric scooters; this paper will look at the impacts that the increasing usage of personal mobility devices are having on the Australian insurance market through a comprehensive review of existing literature and comparison of those findings against local data. As part of this process, we have focused on the different injury frequency and severity experience in a selection of different Australian states and analysing the different factors that may influence them. This includes incorporating the different operating environment that personal mobility devices exist in, including regulation and road rules such as licensing, helmets and speed limits, road infrastructure and insurance regulations and scheme design (i.e. Compulsory Third Party, Workers Compensation).

Following a period of relative favourability with positive customer growth and soft claims experience, Private Health insurance is encountering a new period of risk and opportunity. This risk is driven by increasing cost pressures from hospitals, providers and for customers, the continuing aging of the population, as well as a greater opportunity to contribute to better outcomes from the private health system as a whole. In this context, the Minister and Department of Health are working with insurers, hospitals and other stakeholders to develop a new reform agenda that supports future sustainability including for mental health, maternity care, as well as increased productivity through new models of care and incentives better aligned to more effective outcomes.

What are the perspectives of Health Appointed Actuaries in this changing landscape – is the industry and Government focussing in the right areas, or where should we change to support better outcomes? What role are actuaries playing in this sector, and is that enough or can we have more of a positive impact?

Economic losses from natural hazards continue to grow faster than insured losses, placing increasing pressure on household balance sheets, corporate resilience, and public finances. In Australia, this dynamic is particularly acute given high catastrophe exposure, constrained public budgets, and growing expectations on governments to act as insurers of last resort. This session examines how the protection gap can be addressed not through a single product or policy, but through effective public–private partnership (PPP) structures that align information, risk solutions, and capital across the economy. Drawing on international experience and reflecting on the Australian landscape, the session explores how governments, insurers, and capital markets can work together to improve risk understanding, allocation, and financing.

The discussion is framed around three practical gaps that continue to limit progress: an information gap, a solutions gap, and a funding gap. Better data, modelling, and exposure ownership enable more informed decisions and targeted prevention. A broader toolkit — spanning traditional insurance, structured risk transfer, and alternative reinsurance mechanisms — allows risks to be managed more strategically rather than accumulating on public balance sheets. Finally, access to diversified sources of capital, including insurance-linked and alternative capital, expands capacity and improves affordability. Using examples from global government pools and public-sector risk programmes, the session focuses on how PPPs function in practice, what design choices matter, and how these lessons can inform Australian debates on resilience, funding, and risk sharing. The objective is practical: to move from analysis to execution, and from fragmented responses to coordinated solutions.

Accurately measuring and modelling the dependence between risks is crucial in many actuarial applications. For example, the level of dependence between losses arising from different lines of insurance business directly influences diversification benefits. Hence, dependence modelling is an essential step in setting appropriate capital levels and ensuring solvency (see, e.g., Avanzi et al., 2016).

Current industry practice relies heavily on Pearson’s correlation coefficient, which has well-known limitations. Correlation captures only linear association (Embrechts et al., 2002), meaning two variables can be dependent yet have zero correlation. It is also restricted to measuring pairwise associations (McNeil et al., 2015), and many actuarial models involve more than two random variables. Because pairwise dependence properties do not naturally extend to the multivariate case, the sole use of correlation is intrinsically limiting. For example, correlation will fail to identify a situation where variables are pairwise independent but mutually dependent. This is problematic because many fundamental results − including Central Limit Theorems (CLT) which are pervasive in many applications − often fail under sole pairwise independence (e.g., Avanzi et al., 2021; Boglioni Beaulieu et al., 2021).

In this paper, we explain how distance correlation (Székely and Rizzo, 2009) and Hellinger correlation (Geenens and Lafaye de Micheaux, 2020) address the aforementioned issues and can hence be used as a powerful tool in actuarial applications. Those measures are defined on interpretable distances and are remarkably zero if and only if two variables (or vectors in the case of distance covariance) are independent.

A number of extensions of those measures have been developed in the statistical literature, and we discuss how these can be applied in actuarial practice. For example, joint distance covariance (Chakraborty and Zhang, 2019) captures mutual dependence among more than two random vectors (e.g., multi-line capital aggregation). Auto- and cross-distance correlation (Zhou, 2012) can be used in time series analysis (e.g., claims or mortality), while partial distance covariance (Székely and Rizzo, 2014) can help test conditional independence or guide variable selection (e.g., pricing models).

These new tools offer a robust framework for detecting and modelling nonlinear and higher-order relationships in data, and can help strengthen the reliability of actuarial applications. As global risks become increasingly interconnected − for example, through the effects of climate change − dependencies between risks are likely to strengthen and become more complex. The paper takes particular care in illustrating these concepts in an actuarial context and aims to provide a practical toolkit to those actuaries who may wish to use them in their own practice.

Australian private health insurance generates a rich trail of behavioural and transactional data—quotes, sales, upgrades, downgrades, claims, and life-stage transitions such as adding a partner or children. Yet much of this insight remains underused in shaping products and policies that genuinely reflect consumer needs and value.

This presentation demonstrates how actuarial teams can integrate customer-centred thinking into product design by analysing these behavioural signals through the lens of 3–4 illustrative customer personas—each reflecting a segment commonly targeted by health insurers’ sales and marketing teams. Using de-identified insurer data, we trace how product selection, claim patterns, and policy changes reveal unmet needs, misaligned incentives, and opportunities for better benefit design and communication.

We will also critically examine the current hospital product classification tiers (Gold, Silver, Bronze, Basic) and suggest opportunities for policymakers from an actuarial perspective.

Ensuring fairness in insurance pricing has become a central concern for regulators, insurers, and the public. Existing approaches often rely on descriptive disparity measures or significance testing without explicit reference to fairness concepts or acceptable tolerances, leading to fragmented practices and ambiguous conclusions.

We develop a unified framework that grounds fairness testing in classical statistical inference, providing clear definitions, reproducible methods, and auditable protocols.

Our framework advances fairness testing along four key dimensions. First, we formalise widely discussed fairness criteria that can be potentially applied to insurance pricing contexts as hypotheses on identifiable estimands, making fairness notions statistically precise. Second, we introduce decision rules that incorporate explicit tolerance thresholds, reflecting regulatory standards in practice. Third, we propose inference procedures, bridging actuarial practice with fairness guarantees. Fourth, we design a quote-audit protocol that specifies how to collect, test, and validate fairness claims in a manner that is transparent and replicable.

By combining inferential rigour with regulatory practicality, our approach delivers a coherent methodology that both regulators and insurers can implement. The framework integrates uncertainty quantification and audit design into a single pipeline, ensuring that fairness tests are interpretable, reproducible, and statistically robust. Beyond pricing, the principles can be extended to approval outcomes and process measures, providing a flexible foundation for responsible insurance analytics. The result is a regulator- and insurer-ready toolkit that aligns fairness testing with statistical guarantees and supports the broader goal of accountable and transparent insurance practices.

Rapid economic transformation, driven by forces such as AI, demographic change and evolving societal expectations, is reshaping how sectors deliver value. Injury and disability schemes sit at the intersection of social sustainability, economic productivity and human wellbeing — and actuaries are deeply embedded across this ecosystem, working for many of the stakeholders in this ecosystem and also for the various consultants supporting stakeholders. Generative AI, with its capacity to synthesise data, tailor content and streamline decision-making, presents a pivotal opportunity to recalibrate how these schemes operate and contribute to the next economic chapter.

This presentation draws on collaborative research led by the GIPC Injury and Disability Working Group, with support from the Young Data Analytics Working Group (YDAWG) of the DSAIPC. Through surveys and consultations with scheme actuaries, administrators and digital strategy leads, we map current and emerging uses of generative AI across Australian, NZ and Canadian injury and disability schemes, particularly citing case studies and pilots schemes are deploying. Further, we consider what future advancements in AI may look like for these schemes

In exploring this evolution, key focus areas include:
Scheme Processes
How schemes are deploying or trialling generative AI in real-world contexts — for example, automating claims correspondence, summarising medical reports, enabling customer service interactions, and improving triage for psychological and complex injuries.
Operational Benefits
Ways AI is increasing internal efficiency, easing administrative burdens on case managers, and enabling faster, more informed decisions, particularly in high-volume or high-complexity areas of claims management.
Fraud Detection and Analytics
How generative AI and natural language models are informing predictive analytics, detecting anomalies, and synthesising disparate claim data to support governance and financial sustainability.
Challenges and Guardrails
Consideration of risks and limitations, including data privacy, ethical use, regulatory expectations, bias in large language models and the essential role of human oversight.
Actuarial Involvement and Leadership
Actuaries work across centrally managed and privately underwritten schemes, regulators, consultants, case management providers and rehabilitation specialists. They are increasingly shaping AI strategy — from designing use cases and assessing data readiness to evaluating performance and fairness. This moment offers a uniquely actuarial opportunity to apply rigour, influence governance, balance financial and social outcomes, and guide AI-enabled redesign of scheme models.
Future Outlook
We explore how early successes and emerging lessons, both within schemes and in broader industries, can help actuaries and scheme leaders design more adaptive and client-centred models that support early intervention, enable tailored recovery pathways, and improve data sharing across compensation, health and employment systems — enhancing both social and economic resilience.
This presentation will equip injury and disability scheme professionals with a balanced understanding of how generative AI can be further leveraged to move its use from beyond hype, to become a powerful enabler of sustainable, client-centred schemes. It will also highlight key opportunities for actuaries to lead in this space, by applying actuarial rigour, advancing ethical data use, demonstrating strategic foresight and contributing to the long-term social and financial sustainability of schemes.

Rising disability care costs in ageing societies have intensified discussions on the efficient design of publicly funded systems. These systems vary from individualised, need-based funding to standardised, means-tested packages, differing in support intensity and incentives to invest in health, function, and financial planning. Evidence on whether an individualised model preserves function, economic security, and delays reliance on intensive care remains limited. This study uses a quasi-experimental design to provide causal evidence of the impact of early access to individualised disability support on health and economic outcomes of survival and mortality, retirement savings, and timing and intensity of aged care entry.

Australia provides a rare natural experiment: individuals under 65 with disability may access the National Disability Insurance Scheme (NDIS), which offers flexible, need-based funding. After 65, new applicants are eligible only for standardised, means-tested aged care packages. This age cutoff creates a clear contrast in support intensity, flexibility, and incentives for maintaining health, function, and financial independence. We exploit the age-65 threshold in a fuzzy regression discontinuity design with the staggered NDIS rollout, implementing a difference-in-discontinuities framework. This approach enables credible causal identification in a setting with dynamic, long-term outcomes and staggered policy exposure, a combination rarely studied in health economics.

Using population-level linked administrative data containing social services, census, and tax records (PLIDA) from 2011 to 2024, we follow individuals just below and above the cutoff across outcomes. Our study population draws from 1.4 million individuals aged over 65 accessing care support. Our treatment group consists of 39,000 NDIS recipients currently aged over 65, while our primary control group consists of 275,000 participants of the Home Care Packages Program.

We compare discontinuities across granular regions and rollout periods to isolate the effect of early, flexible support from age-related trends and confounders. We also examine heterogeneity by baseline functional capacity, gender, and disability type, and test robustness to alternative bandwidths, placebo cutoffs, and pre-trend specifications.

This research informs Australian policy on life-course investment and provides insight into key normative trade-offs in social care reform, including balancing upfront investment with downstream savings, fairness with fiscal sustainability, and flexibility with standardisation.

Fundamentally, actuaries are asked to put a number to uncertain future events. In recognition of future uncertainty, we often provide a range of reasonable outcomes. As part of standard valuation work, actuaries will present a central estimate alongside a metric for uncertainty, for example, a quantile of the loss distribution in the form of a risk margin or a risk adjustment.

For the purposes of assessing experience, extracting useful operational insights and updating assumptions, there is often much focus on the central estimate and less emphasis placed on the distributional information. Common methods of hindsight analysis tend to compare actual outcomes to point predictions rather than an explicit range from the loss distribution. Actuarial judgement is then often used to justify the materiality of differences and determine whether changes in methodology/assumptions are warranted.

We propose a simple framework to supplement hindsight analysis and judgement using the (often already available) forecast distribution. This approach provides valuable additional information by granting a richer understanding of the range of possible outcomes. It inherently accounts for the variability of the predicted estimate and allows quantification of how outlying each new observation is, based on the characteristics of the line of business such tail length, portfolio size and large claims volatility. In this paper, we will demonstrate how this information can be readily constructed from work already done in deriving the risk margins/adjustments.

Our proposed framework applies established metrics from the field of probabilistic forecasting to assess the distributional performance of actuarial valuations, through a process that integrates easily with standard actuarial valuation pipelines. This information could be used in a number of practical applications, such as experience monitoring, a trigger for reassessment of risk margins/adjustments or as a tool to evaluate the appropriateness of the current valuation assumptions.

By integrating probabilistic forecasting into actuarial valuations, actuaries can move beyond mean-based comparisons towards a more robust examination of emerging data. This will strengthen evidence-based decision making, and will provide a more holistic, quantitative view of how our models are playing out in practice.

In an era of data-driven decision-making, organisations face a growing need to balance agility with control. This paper presents a practical and strategic approach to uplifting enterprise data maturity through federated data governance, self-service reporting enablement, and the adoption of modern architectural constructs such as data mesh and medallion architecture. Drawing on real-world implementation within a regulatory body overseeing portable long service leave, we explore how decentralised ownership, standardised modelling, and layered data pipelines can transform reporting and analytics capabilities across the enterprise.

The federated governance model empowers business units to act as data product owners and stewards, while a central data team provides reusable assets, quality standards, and semantic consistency. This approach enables scalable self-service reporting through enterprise-grade Power BI templates, governed semantic layers, and a shared understanding of metrics. By embedding roles such as data custodians and stewards, and aligning them with business accountability, the organisation fosters a culture of trust, transparency, and shared responsibility.

We also demonstrate how medallion architecture — with bronze (raw), silver (cleansed), and gold (business-ready) layers — supports modular, auditable, and performant data pipelines. Combined with data mesh principles, this architecture decentralises data ownership while maintaining interoperability and governance, enabling rapid delivery of insights without compromising quality or compliance.

This paper offers a blueprint for actuaries, data leaders, and regulators seeking to build resilient, scalable, and future-ready data ecosystems. It highlights how governance, architecture, and culture must evolve together to support advanced analytics, AI readiness, and strategic foresight in complex organisations.

In Australia, community rating ensures everyone pays the same premium for the same level of private health cover, regardless of age, gender, health status, or claims history. To support this, risk equalisation re-distributes part of the claims costs from insurers with higher-than-average benefit payouts to those with lower-than-average payouts, balancing financial impact of covering higher-risk individuals. While community rating and risk equalisation are here to stay, health analytics continues to play a vital role for Private Health Insurance (PHI) in Australia. This session will showcase practical examples of how health analytics drives efficiency and value across the PHI landscape.

For example, behavioural models may be integrated with data analytics to allow insurers to better understand the drivers of health-related behaviours among their members. With increasing amount of data collected outside of the insurance itself, such as from wearables, affiliated clinics and sponsored events, patterns and triggers may be monitored and analysed to encourage toward healthier behaviours. Predictive modelling also helps to identify claims leakage and informs the development of more robust claims rules to prevent future losses. Additionally, by forecasting claims volumes and complexities, predictive modelling supports more effective workforce planning, ensuring resources are aligned with expected demand. Also, analytics-powered lifetime value (LTV) models uncover attributes of high-value policies, guiding acquisition and retention strategies. Insurers can embark on a data-driven approach to tailor marketing and engagement efforts more effectively.

Medical indemnity insurance, also called medical malpractice insurance, is a specialist professional line of general insurance which plays a crucial role in supporting the Australian community. Without MI coverage, healthcare professionals would find it challenging and, in many cases, prohibitive to operate under the downside risk of potentially large lawsuits when practising their craft and delivering essential services for Australians.

In this paper, we will provide an update on the state of the MI market in Australia and overseas. We will highlight key risk and market trends, drawing on publicly available data and resources as well as the authors’ professional experience with MI. The intention is to provide a comprehensive review of the last decade or so in the market and how it continues to evolve alongside medical and legal sector trends. We will identify the key issues and opportunities likely to shape the MI market over the years ahead.

Actuaries working with MI will directly benefit from a concise summary of the MI sector, which is generally not publicly available or regularly updated given the specialised nature of MI. General insurance actuaries more broadly will be exposed to a class of insurance which most do not deal with on a regular basis, which will hopefully spark greater interest and further expertise by the actuarial profession.

The healthcare sector was under the spotlight during the recent global pandemic and healthcare crisis, and given the important role MI plays in the sector, this presentation will also be of significant interest to actuaries outside general and health insurance. As the nature of medicine and the legal environment continues to rapidly evolve, MI provides a solid area where actuaries can help society navigate risk and continue delivering in the public interest.

Whilst Excel has remained a flexible and handy program for over 750M users, including actuaries and E-sports participants, recent advancements in data science and AI has seen an increasing adoption of Python – a popular functional language amongst developers and data professionals. As a practitioner familiar with both tools, I identify the following shortcomings of Excel-heavy data processes, where Python may offer an advantage: • Row and column limit on ingesting data • Scalability challenge with respect to linking or managing data across multiple Excel workbooks, or performance significantly slowing with large datasets The workshop / demonstration will focus on practical data transformation tasks where Python may deliver greater efficiency (and #VALUE!). Examples include but are not limited to the following: • Search of common data points between multiple columns • Filtering for multiple values in a column • Joining or concatenation of multiple datasets • Fast pivot-tabling • String and date manipulation • Window functions The workshop / demonstration will be based on publicly available datasets, such as insurance datasets from the book ‘Actuarial Data Science with R’ (CASdatasets).

In the rapidly evolving world of non-life insurance and in the middle of the AI innovation explosion, the fusion of finance, operations, underwriting and actuarial science is transformative. As an Actuary, Finance Director/CFO, and former TED speaker with over 20 years of speaking experience, I will explore how actuaries can redefine financial strategies to drive strategic value and achieve success.

Through compelling case studies, discover how actuarial expertise in reserving, capital allocation, risk management, and profitability enhancement is reshaping the insurance landscape. These stories highlight actuaries as catalysts and not inhibitors of change and growth.

We will discuss how actuaries leverage their skills in reserving, pricing, underwriting, operations, and performance monitoring to provide insights that fuel business growth and resilience. By integrating their knowledge with financial metrics and narrative building, actuaries empower organizations to navigate challenges and seize opportunities.

This talk will inspire actuaries to embrace their role as strategic partners in the non-life insurance domain, fostering innovation and resilience. Attendees will gain a renewed understanding of how actuaries are shaping the future of non-life insurance for sustainable success.

Join me to explore the transformative potential of actuarial science in non-life insurance and embark on a journey toward strategic excellence.

As the rollout of AI continues to grow, there is continued interest in transferring ideas to see how deep learning models can be applied in new ways. For actuaries, there is an ongoing opportunity for new models to handle multivariate prediction problems where there are important dependencies between outputs. Examples include: claims reserving (estimating a suite of accident and development period combinations), pricing (where different risks are estimated and combined into a risk price), microsimulation (where many different characteristics are simulated over time in a consistent way) and DFA models (where different parameters will evolve over time to determine capital positions, with strong correlations across measures).

Diffusion models are a common deep learning approach, particularly in image generation. Starting with random variation, a trained model will successively ‘de-noise’ a dataset until it resembles something coherent. In many cases they outperform other model structures, including in a small number of research papers testing applicability to tabular data. Our paper will apply diffusion models to key actuarial problems. The work will build on our 2025 summit paper on deep learning models – we test an entirely new way to set up and train the deep learning model (a diffusion framework), while also significantly broadening the range of applications of our work. In addition to microsimulation and time series problems previously explored, we will test diffusion models’ applicability to reserving contexts. The paper will cover simulated and real-life data examples.

The models represent an important contribution, with the potential to massively simplify the training of complex systems. To the author’s knowledge, this is the first time such diffusion models have been applied in an actuarial context.

As actuaries, we rely on observational data to build models, understand historical patterns and tell stories, those datasets show us what has happened, but rarely tell us why things happen. The why matters because it pin points to action and consequence, it enables decision makers understand causal effects of changes rather than blinded by correlations, and it provides robust policy evaluations.

Understanding the why, or the cause and effect is challenging, there are techniques and methods to illuminate the hidden causal relationship, randomized controlled trial (RCT) is often considered as the gold standard for uncovering causality, as it directly addresses the problem of endogeneity by ensuring treatment assignment is exogenous, while other causal methods such as differences-in-differences, regression discontinuity, or instrumental variables arguably suffer from the endogeneity problem to various degrees.

RCT, as a method of field experiment, generates meaningful data, sharper insights and business guidance beyond observational data can provide, it adds to actuaries toolkit and redefines our understandings of why things happen. This presentation will cover common causal inference methods, endogeneity problem, motivations and pitfalls of RCT and a case study of field experiment, which hopefully will inspire more actuaries to explore and embrace the spirit of experimentation.

With shifting demographics and escalating care costs placing unprecedented strain on the national budget, Australia’s health economy is at a pivotal inflection point. The convergence of rising chronic disease, workforce shortages, and rapid digital transformation—driven largely by AI and data innovations—demands a new strategic approach. The pandemic exposed both the fragility and impressive adaptability of our health infrastructure, underscoring an urgent need for reform. As Australia seeks to lift productivity and achieve fiscal and social sustainability, healthcare reform must be a cornerstone of The Great Recalibration powering the next economic chapter.

Actuarial science offers a unique capacity to translate complex population health risks into long-term fiscal and operational insight. This paper explores how state-of-the-art actuarial analytics—including dynamic risk-adjustment, scenario simulation, AI-enhanced forecasting, and value-based funding models—can reshape decisions on healthcare funding, prevention, and care delivery. We highlight the integration of prevention economics into actuarial frameworks to quantify the productivity dividends of healthier populations.

Drawing on comparative insights from Australia, the U.S., and India, the session illustrates how diverse systems embed value-based care principles within their financing frameworks. Case studies demonstrate how actuarial modeling can guide optimal resource allocation across prevention, primary, and aged care—quantifying not only costs but future capacity, workforce resilience, and system adaptability.

Ultimately, the recalibration of the health economy is more than budgeting; it is a test of societal foresight. By aligning actuarial models with public health and economic objectives, funding systems can be redesigned to reward long-term outcomes over short-term expenditure, helping Australia power a more sustainable and resilient economic future. This session provides a practical framework for actuaries to become central architects in this recalibration, equipping them to convert health data and AI-driven insights into strategic intelligence.

Shakespeare believed that love and war were so important and overwhelming that people should be excused for acting in their own best interest. But general insurance is neither a battlefield nor a romance - it's a relationship built on trust, where fairness must prevail over self-interest. In his October 2024 dialogue paper, “Fairness in Insurance”, Ian Laughlin challenged insurance Boards to elevate customer fairness as a strategic priority. This paper shifts the lens to explore the critical role general insurance actuaries can—and should—play in delivering fairer outcomes.

Laughlin recognised that Boards occupy a unique vantage point for ensuring fair customer treatment, and identified fairness as a "natural area of interest" for actuaries. But what are the obligations and rules of fairness which apply to general insurance actuaries? This paper examines what practitioners need to know about fairness and what they could, or should, be doing to achieve this objective in practice.

While Laughlin noted that Boards benefit from sitting "above the fray" with greater objectivity than management, this paper argues that general insurance actuaries are equally well positioned—if not better placed—to balance competing interests. Working in the front lines where pricing, underwriting, and claims decisions are made, general insurance actuaries can serve as vital arbiters of fairness, translating principle into practice where it matters most.

Risk is evolving at an extraordinary pace, well beyond what traditional frameworks can capture. Geopolitical instability, climate volatility, technology and cyber risks and emerging health challenges are redefining what customers need protection against, and what insurers can sustainably cover. Through four focused lightning talks, hear from Alix Pearce, Insurance Council of Australia, and Rade Musulin, Finity Consulting who are at the forefront of these emerging risk areas. This session will examine the changing nature of insurable risks and what adjustments are needed across products, pricing and risk management. Walk away with concrete strategies for adapting your actuarial practice to meet emerging risk challenges.

Gain insights from speaker Karan Anand, Chief Strategy Officer and Managing Director Australia at Hnry. Rapidly changing consumer expectations and economic transformation demand actuarial insight that anticipates and then moves ahead of the speed of change — designing solutions rapidly, testing ideas fearlessly and communicating impact clearly to executive teams. How do we bring startup agility into actuarial practice while maintaining rigour? Karan shares frameworks from building and scaling innovative businesses. Through real examples of commercial innovation, you'll learn how to run lean experiments, focus resources on what's material and build business cases that resonate with leadership.

Disruption demands the ability to rethink, reset and act. Hear from accomplished actuaries, Adam Driussi, CEO of Quantium, and Nicolette Rubinsztein AM, who have faced defining moments, tough calls, unexpected failures and breakthroughs that reshaped their thinking. They will share the lessons, mindset shifts and practical actions that turned setbacks into stepping stones for their careers. You will leave with stories that can be translated into lessons for today and ways to build resilience and reframe setbacks, to turn challenges into growth. More speakers announced soon.

The actuarial profession stands at a pivotal crossroads. As the insurance landscape evolves and actuarial software continues to advance, the role of actuaries is undergoing a significant transformation. The integration of AI capabilities has further reshaped this role, positioning actuaries not just as technical experts, but as strategic partners in shaping business direction and driving growth. Today, actuaries are increasingly recognized for their contributions to strategic decision-making, working closely with management teams to define growth trajectories and business outcomes. Modern actuarial software, now embedded within broader data and analytics ecosystems, empowers actuaries to unlock the full potential of insurers’ data assets. With enhanced collaboration and smarter tools, actuaries are accelerating speed to market and delivering more impactful, growth-oriented results. In this presentation, we will showcase real-world examples and best practices that highlight how actuaries are leveraging technology to move beyond traditional support functions—and into roles of strategic influence

Boomers and Zoomers may share DNA, but their lifestyles are like night and day, and so are their health outcomes. This presentation explores how generational differences in lifestyle, and environment have shaped not just their culture but their cholesterol, cortisol, and coping mechanisms. Spoiler: one group likes red wine; the other looks for red flags. But what does this mean for life insurance underwriting and pricing? Smoking: from cool to cancelled Boomers grew up in a haze of cigarette ads and smoky bars, leading to higher rates of lung disease and heart issues. Zoomers, raised on anti-smoking PSAs, treat smoking as a social crime but vaping is considered “slay”. While smoking rates have dropped, vaping introduces new unknowns. Are current underwriting assumptions keeping pace? Alcohol: martini’s vs matcha Boomers made casual drinking a cultural norm, now facing consequence like liver issues and high blood pressure. Meanwhile, Zoomers chase “insta-worthiness”, flocking to photogenic cafés for matcha lattes and oat milk everything. The outcome? A steady rise in diabetes across generations. Are we capturing these risks accurately? Fitness: sports vs e-sports Boomers think fitness means chasing a golf ball and calling it cardio, with a side of nostalgia for team sports. Zoomers calling dance challenges “exercise” and pickleball a sport. Boomers may have stronger knees, but Zoomers have stronger Wi-Fi. As screen time rises and step counts fall, what’s our long-term view on health impacts? Mental health: stiff upper lip vs. soft launching your therapist Boomers often ignored mental health, leading to late-stage diagnoses. Zoomers are emotionally fluent and therapy-positive yet overwhelmed by rising rates of anxiety and burnout. With mental health claims on the rise, are legacy underwriting frameworks still fit for purpose? From chain-smoking to chain-swiping, from lighting up cigarettes to lighting up scented candles, everyday habits shape health outcomes. Yet pricing, product and underwriting within life insurance often lag behind these generational shifts. This presentation compares the habits and health impacts of Boomers and Zoomers, and challenges the industry to rethinking assumptions for a changing world.

Australian Life, General and Health insurers have experienced significant uplift in APRA’s prudential requirements for risk management frameworks over the last 15 years, with CPS 230 Operational Risk Management being the latest instalment. For health insurers, who have more recently had to comply with APRA’s capital framework, APRA observed that not all health insurers understand how all elements of the risk management framework are intended to interact. Lessons from economic history can help insurers to better understand how risk management frameworks are expected to operate in practice, specifically the relationship between Board and management in setting risk appetite; and cascading components of the risk management framework throughout an entity. The purpose of this presentation is to highlight some of lessons we can learn from economic history that can assist insurers to enhance the practical implementation of their risk management frameworks.

Australian Life, General and Health insurers have experienced significant uplift in APRA’s prudential requirements for risk management frameworks over the last 15 years, with CPS 230 Operational Risk Management being the latest instalment. For health insurers, who have more recently had to comply with APRA’s capital framework, APRA observed that not all health insurers understand how all elements of the risk management framework are intended to interact. Lessons from economic history can help insurers to better understand how risk management frameworks are expected to operate in practice, specifically the relationship between Board and management in setting risk appetite; and cascading components of the risk management framework throughout an entity. The purpose of this presentation is to highlight some of lessons we can learn from economic history that can assist insurers to enhance the practical implementation of their risk management frameworks.

The CEO of LGI (life, general, and investments) wakes up at 7am, there are already ten burning problems in their inbox. It’s the year of 2026, Mr/Mrs CEO doesn’t live in the stone age anymore, the goal is to solve some of these problems with the help of AI today. This presentation is an interactive, role-play session that places actuaries inside a realistic, fictional company to showcase practical AI applications in an industry context. In this session, we will act out and explore several use cases of AI in the workplace, demonstrating both the opportunities and challenges actuaries may encounter when integrating AI into everyday work. Beyond showcasing applications, the session will dive into practical issues surrounding AI implementation and governance, examine the differing perspectives of executives and boards toward AI adoption, and highlight the importance of human judgment and accountability in an increasingly automated world. Use case examples (subject to refinement): 1. Live claims experience analysis - contrasts traditional manual review with an AI-enabled approach that rapidly surfaces patterns and anomalies in claims experience. The side-by-side view highlights how AI shortens time to insight and improves decision accuracy. 2. Motor insurance pricing - explores how shifts in vehicle technology and street design change crash frequency and severity, and how AI can quantify these effects in quick “what if” scenarios so insurers can adapt pricing and underwriting with confidence. 3. Investment sentiment analysis - AI distils signals from news, earnings calls, and social media to surface early, actionable sentiment. This use case offers a forward-looking perspective on AI’s role in investment decision-making. This interactive session brings rigorous actuarial thinking to life in a fun, memorable way through practical AI demos that will spark discussion on reliability, governance, and the human–machine collaboration essential to a brighter future for the industry. Goal of our session: Help Actuaries understand and internalise concrete examples of AI solving real business problems in the way these tools are being used today. Attendees will leave with: • A clear understanding of when AI adds value vs when human judgement is critical, • Useful frameworks can be used for thinking about AI governance, and • A realistic view of implementation challenges

Agricultural systems are becoming progressively vulnerable to climate change, which exacerbates yield variability and undermines the efficacy of traditional agricultural risk transfer mechanisms. In Australia, limitations of traditional multi-peril crop insurance have sparked ongoing policy and regulatory interest in broadening mechanisms for agricultural risk transfer. Within this context, the Australian Government’s Future Drought Fund (FDF) has funded pilot Weather Index Insurance (WII) initiatives that base indemnities on independently verifiable climate indicators, such as rainfall and soil moisture, rather than on field-level loss assessments. Nevertheless, a critical limitation of WII is its exposure to temporal basis risk, whereby misalignment between index measurement windows and crop-specific phenological stages weakens the correlation between index outcomes and realised yield losses. Hence, temporal basis risk potentially undermines product credibility and uptake (Ntukamazina et al., 2017; Tappi et al., 2023; Zou et al., 2024). Temporal basis risk may arise in traditional WII design if the weather index is aggregated over the entire growing season, obscuring these stage-specific sensitivities. An effective WII product design should reflect that each physiological stage of crop development exhibits distinct sensitivities to specific climatic stress factors (Dalhaus et al., 2018; Nleya et al., 2013). For instance, severe flooding during the vegetative stage may lead to widespread plant mortality (Fletcher et al., 2023; Wu et al., 2016). However, subsequent favourable weather conditions during later stages of the crop growing season may contribute to an overall positive weather index value. Another source of temporal basis risk is the simplified assumption that the duration of growing stages remains constant over time. Thus, if the index is not aligned to the time periods of these variable growth stages, then the index may miss important weather conditions that impact the crop yield. This research proposes a novel hybrid WII design with a staggered index and payout structure aligned to crop growth stages, rather than a single aggregate index and payout at harvest. The hybrid WII product has been evaluated using soybean data from the US Corn Belt Region. The soybean growing season is divided into three phenological stages: vegetative (VE to V6), pod & seed set (R1 to R4), and seed fill (R5 to R6). Each of these stage durations are determined using a combination of soybean maturity group classification, predicted planting dates, and accumulated growing degree days (GDD). Weather indexes are developed for each crop growth stage using machine learning models to align with the specific physiological factors that are sensitive to each stage. The benefits of this hybrid model are outlined below. 1.Dynamic Crop Growth Stage Timing: The variable timing of the crop’s actual growth progression is considered, ensuring that payouts are directly aligned with observed weather conditions during the phenological development. 2.Capturing Sensitivities by Growth Stage: The weather index for each critical crop growth stage accounts for the crop's physiological sensitivities to climate conditions during that growth stage. 3.Early Season Loss Compensation: Early growing season payouts can provide liquidity to farmers to invest in replanting. The opportunity to replant improves crop yield outlook. These funds can also be invested in futures contracts to lock in prices and manage price risk. The stacked ensemble model was utilised to develop weather indexes for each crop stage with significant risk reduction, measured with the spectral risk measures, of 20% (Iowa), 17% (Illinois), 15% (Ohio), and 14% (Indiana) during the 2003 season which was impacted by severe drought. SHAP plots were employed to verify that the model predictors correspond with crop sensitivities at various growth stages. Overall, this hybrid product design with staggered payments mitigates temporal basis risk by more precisely aligning indemnification with crop phenology and vulnerability to adverse weather conditions. By addressing a fundamental obstacle in the adoption of WII, these advancements enhance the credibility of WII as an effective risk management instrument and reinforce the resilience of the agricultural sector. Although the empirical application of this study is grounded in the US due to the availability of high-resolution agronomic and climate data, the proposed framework is readily transferable to other climate-exposed agricultural systems, including Australia. The Australian agricultural sector has ongoing policy and regulatory interest in exploring data driven WII and increasing farmer uptake. Hence, these considerations position the proposed framework as a transferable methodology to support agricultural resilience in climate exposed markets such as Australia.

Across many low and middle-income countries (LMIC), the supply of actuaries is limited or non-existent. Actuarial skills are essential for the development of the insurance sector, which is one of the key pillars in building a sustainable and resilient society. Additionally, integrating actuarial expertise into government functions can lead to better economic and societal outcomes. In this session, we will explore how actuaries contribute to the UN Sustainable Development Goals, examine the different stages of actuarial profession development, and highlight the challenges faced at each stage. We will share insights and identify characteristics of different phases, key drivers and the critical elements required to ensure sustainable growth of the profession. UNDP-Milliman Global Actuarial Initiative (GAIN) is a pro-bono program which started in 2022 looking to build the actuarial profession in LMIC. Since the inception of the program, GAIN has involved over 150 volunteers contributing over 27,000 pro-bono hours across 16 countries. Drawing from practical experience, we will also discuss key interventions and prerequisite steps needed to progress to the next phase and highlight the critical role that diverse stakeholders including actuaries from diverse backgrounds can play in actively supporting this effort. Building and strengthening the actuarial profession requires deliberate and strategic efforts, underpinned by coordinated multi-stakeholder engagements ranging from regulators to educational institutions to actuarial associations. The presentation will look into the following aspects: (1) increasing the supply, expertise and experience of actuaries, (2) increasing the demand for actuaries and key roles for them, and (3) creating a facilitative environment, along with resources and support so that actuaries thrive in their roles and provide increasing value to their employers and the broader economy. Actuaries rely on strong regulatory frameworks to practice effectively, and through their technical expertise and industry engagement, they also play a vital role in shaping and strengthening these frameworks. This session will explore: 1. The challenges faced by these LMICs in building in-country actuarial capacity and its consequences. 2. The key characteristics of different development phases of actuarial profession progressing from the nascent phase to the foundational phase, followed by the integration phase, professional maturity phase, and culminating in strategic leadership and global engagement. 3. The key learnings from GAIN in supporting the development and strengthening of actuarial capacity in selected countries across three continents (including Nepal, Nigeria, Jordan and Colombia). The session will also explore the ways in which volunteers can contribute to advancing such effort.

Cyber risk is recognised by many as the top global risk in 2025, with data breaches, ransomware attacks, and system compromises posing significant threats to all organisations and customers. But despite their growing frequency and severity, quantifying their financial and operational impact remains a major challenge. Scenario testing is a foundational tool – not only required for insurers’ capital adequacy assessments, but also to help all organisations visualise and prepare for their key risks. However, most insurers lack a robust methodology for cyber scenarios, often relying on subjective estimates without clear justification or consideration of the potential key exposures and costs. Through the perspectives and experience of both capital management experts and cyber risk experts, this session will explore: • Analysis and learnings from recent major cyber events and their implications for capital, business continuity, and future best practices in cyber risk management. • Cyber risk quantification and how it relates to cyber risk exposure quantification, investment prioritisation, insurance coverage, and the business’ operational resilience. • The range of maturity of how cyber scenarios are currently done by industry, including examples of better practice. • A comprehensive framework for considering different cyber disruption scenarios for likely future cyber incidents. Join us to understand how to evolve your understanding and approach to managing cyber risks.

In recent years, “quantum computing” has jumped from lab notes to board-level disclosures and government MOUs. HSBC reported a 34% lift in trading performance. PsiQuantum, worth $10 billion+ after a recent capital raise, has received $940 million of funding from the federal and Queensland government to build a quantum computer in Brisbane. And, the recently-inked Tech Prosperity Deal between the US and UK listed 7 focus areas for collaboration to “Secure Quantum Advantage.” This session will get you ahead of the curve: able to read quantum headlines sensibly, separate signal from marketing, and answer the inevitable “should we care yet?” with confidence. We’ll get there via a short, non-technical tour of classical computing (including the abacus) and a simple, memorable picture of what makes quantum different so you can talk about it without new jargon. As with previous waves of emerging technology such as data and AI, early interest in quantum computing is accompanied by heightened expectations. The challenge for organisations will be distinguishing genuine progress from short-term hype and assessing when meaningful opportunities will emerge.

Workers compensation reforms across Australia are redefining the boundaries of mental health claim eligibility, particularly through stricter criteria and revised impairment thresholds. These changes are reducing access to compensation for individuals with psychological injuries, shifting the financial and operational burden to life insurers—especially within group and individual disability income products. This shift from compensation to coverage introduces new challenges for actuaries. While the immediate financial impact may appear modest, emerging behavioural trends—such as increased legal contestation, longer claim durations, and altered claimant pathways—suggest potential for cost escalation. These developments raise critical questions around pricing assumptions, reserving methodologies, and capital strategies, particularly in the context of mental health-related claims. This session will explore the actuarial implications of these reforms, including the difficulty of quantifying latent risks, managing uncertainty in claim trajectories, and navigating data limitations. It will also consider mitigation strategies such as benefit design adjustments, enhanced analytics, and cross-sector collaboration. Actuaries across life, general, and health domains are uniquely positioned to respond to these changes. By re-evaluating traditional frameworks and embracing a more holistic view of mental health claims, the profession can help shape more resilient insurance and compensation systems.

As Australia embarks on its first year of mandatory climate-related financial disclosures, the actuarial profession stands at the forefront of a transformative shift in corporate transparency and sustainability governance. This session explores the inaugural wave of sustainability reports published under the new climate reporting regime AASB S2, with a comparative lens on insurers who have led the way. Drawing on benchmarking insights from the Australian sustainability reports and evolving regulatory expectations, we will unpack: • Implementation of AASB S2: How insurers have interpreted the standards, including common themes and divergences in disclosure practices • Scenario Analysis & Risk Quantification: Challenges in scenario modelling, data readiness, • Governance & Strategy: ESG governance structures, assurance requirements, and integration of climate risk into strategic planning • Metrics & Targets: Approaches to climate modelling metrics and GHG emissions reporting • Operational Considerations: Data challenges, technology solutions • Industry Best Practice and Next Steps: How leading insurers are adapting risk frameworks and operating models to meet climate disclosure demands and how we expect this to evolve This presentation will highlight the actuarial profession’s critical role in supporting insurers through this transition—bringing analytical rigour, risk modelling expertise, and forward-looking insights to the forefront of climate accountability. Aligned with the conference theme “The Great Recalibration: Powering the Next Economic Chapter”, we’ll explore how actuaries can help embed resilience, accountability, and sustainability into Australia’s financial systems.

Reputation risk has been difficult to quantify because observable data are sparse, lagging, and noisy across multiple stakeholders. Impacts arrive through indirect pathways that are hard to attribute, such as attrition, funding costs, regulatory posture, and franchise value. Organisations have therefore relied on narrative, subjective, and reactive approaches anchored in communications rather than risk measurement, with little ability to define frequency and severity, test counterfactuals, or backtest decisions. The result has been limited actuarial input, weak governance, and few objective metrics that link stakeholder reactions to capital or performance outcomes. Generative AI changes this landscape by converting unstructured sources into analysable datasets and by simulating plausible human responses to events and management actions. News, regulatory filings, customer complaints, social media, and earnings calls can be transformed into structured signals for prediction and monitoring. Within the actuarial control cycle, these capabilities enable clearer problem definition, data acquisition, model building, decision implementation, and feedback with auditable artefacts. Generative AI is therefore a paradigm-shifting tool that opens reputation risk to actuarial practice. Actuaries’ training in ethics, stakeholder trade-offs, and risk governance positions them to advise on model validity, bias mitigation, and the optimisation of communication and remediation strategies. The paper and presentation provide practical case studies and simulations that actuaries can apply immediately in banking and other sectors. Examples illustrate how to parameterise event frequency and severity, model stakeholder reactions, and translate results into loss channels and management playbooks. All code, schemas, and tools used for event structuring, simulation, and stress testing will be released as open source to support replication, peer review, and adoption across the profession. I will present in person. The session will combine presentation slides with active audience participation. Three pre-briefed attendees will join me on stage to role-play distinct stakeholders and be interviewed on how they would respond to a hypothetical scenario. Later, the audience will vote on the most effective mitigation strategy across several scenarios, and we will compare that crowd judgement with the model’s recommended strategies, highlighting where communication, governance, and optimisation meaningfully change predicted outcomes.

We all face ethical dilemmas from time to time. The “right response” to an ethical challenge can change with a subtle twist or turn in the situation and the information available. In addition, different people will have different views on the “right response”, and when combined with topical issues for actuaries and professionals, we have discussion ripe for exploration at the All Actuaries Summit 2026, using a presentation model drawn (copied shamelessly) from Geoffrey Robertson’s Hypotheticals. This session will include a number of brave panellists, who are prepared to respond to an evolving situation and offer their thoughts on the ethical dilemmas that arise. This is not a scripted performance, and panellists will be provided with very little scenario information ahead of time – so will have little or no opportunity to prepare their positions. The purpose – and value – of the session is to: • Draw attention to the Code of Conduct, and the assistance available to members when faced with ethical dilemmas • Remind the audience that ethical dilemmas can lurk in much of what we do • Reassure members that these things are hard – but worth working through • Encourage discussion with peers • Entertain – not to make light of the situation, but to encourage engagement with it.

The 2025 Nobel Prize in Economic Sciences, awarded to Joel Mokyr, Philippe Aghion, and Peter Howitt, celebrates a unified vision of innovation-driven growth—where progress arises not merely from technology, but from the cultural, institutional, and competitive forces that allow innovation to flourish. This perspective holds profound relevance for the actuarial profession, especially within private health insurance, where uncertainty, incentives, and social value converge. This talk explores how actuaries can translate these theoretical insights into practice. Drawing on Mokyr’s view of cultural and institutional evolution, actuaries can design insurance systems that reward preventive care, data-driven innovation, and healthy behavioural change. Guided by Aghion and Howitt’s model of creative destruction and endogenous growth, actuaries can foster market environments that encourage innovation while maintaining fairness through mechanisms like dynamic pricing, risk adjustment, and equitable access. Finally, by quantifying and managing the systemic risks of innovation, actuaries can ensure that health systems remain both adaptive and stable amid technological disruption. In essence, this session argues that actuaries are not merely guardians of financial soundness—they are architects of sustainable innovation. By integrating economic theory with health system design, actuaries can help transform private health insurance from a cost-containment model into an engine of long-term societal well-being.

The insurance landscape is continually evolving in response to climate change, shifting demographics, and changing stakeholder expectations. For life insurers, maintaining resilience and achieving financial and social sustainability is paramount. Our speakers will underscore the growing significance of understanding climate change impacts on mortality and morbidity risks. We will present research findings on climate change's influence on mortality and introduce a framework for assessing impacts on morbidity, with a particular focus on mental health. As reporting requirements on climate change become increasingly stringent and quantitative, the industry must enhance its understanding of exposure, impacts, and future scenarios. This presentation will offer recommendations for proactive monitoring of climate change impacts and strategies for risk mitigation. Beyond the direct effects on claims incidence, we will also examine: 1 The broader implications of climate change for life insurers, including client support for financial hardship, sales and marketing strategies, underwriting practices, and claims operations. 2 Key takeaways and recommendations for life insurers, emphasizing the need for enhanced scenario modelling and collaboration with other stakeholders to develop effective solutions to climate change challenges. We look forward to your participation in this critical discussion.

Catastrophe risk makes up approximately 50% of the risk in the Australian general insurance property market and is currently managed via some of the world’s largest reinsurance programs. Imagine a future where: • In the coming season, the chance of a major catastrophe event and subsequent events, is well known. • Portfolios and risk appetites can be shaped on-the-fly in response to regular forecast updates • The price of insurance varies week to month like airline tickets. What does that mean for our customers? For the stability of our insurance market? For reinsurance markets? Over the past 15 years, the emergence of detailed spatial perils data has fundamentally reshaped the general insurance market’s approach to risk-based pricing for weather-related perils, driving significant shifts in affordability and the need for government intervention. Advancements in seasonal weather forecasting, combined with emerging temporal capabilities in catastrophe models, are opening up new opportunities to transform how catastrophe risk is managed. These innovations enable dynamic assessments of catastrophe allowances, capital requirements, and reinsurance needs on a seasonal basis, with potential implications for how these factors are reflected in pricing strategies. The introduction of the next generation of high-definition catastrophe models to the Australian market enables granular pricing across the board, shifting the competitive focus among insurers. Rather than relying on a superior temporally averaged View of Risk, the emphasis is now moving toward a dynamic, seasonally responsive View of Risk. Further, strict and impractical perils definitions are breaking down as closely coupled perils event sets emerge, allowing better representation of losses across storm, flood and other weather events. We explore several use cases for temporal catastrophe risk management, including seasonal forecasting, peril correlations, and policy-level seasonal loss correlations. We examine the implications for dynamic capital management and highlight emerging considerations for regulators, reinsurance strategies, and the pricing of reinstatements. Finally, we consider potential outcomes for consumer pricing behaviour to support a value-based dialogue on the future direction of the industry.

This paper is about a life insurance industry we could build if more failure was encouraged. Life insurers, in contrast to many other industries, have the distinct advantage of being cashflow rich and able to make really long-term bets. Yet we struggle to point to many groundbreaking innovative and successful changes. Many believe that the most recent were the introduction of trauma products, developed in the 1980’s, but since then change has arguably been a series of tweaks or partial adoptions rather than another seismic shift. And every seasoned market participant can tell you a story about how they have once been burned by an initiative or product that sold only a few policies or led to a major loss. This paper explores some of the global life insurance product failures, their hidden stories and their key lessons. It suggests that we actually have a different time lens to work with and regardless of whether the innovations succeed or not, the problems they are solving haven’t changed. It recognises that we can’t have innovation without failure and proposes a metric to be used to encourage, measure and reward failure which has its base the idea that innovating is all about trying to fail.

With GenAI becoming embedded in insurance workflows, actuaries increasingly need to validate, not just build, AI models. How can we trust models we didn’t design, that have inherent randomness and acknowledged limitations? Drawing on case studies of real AI deployments in market, the session will provide practical insights and workflow suggestions. This session will give you a deeper understanding of how language models work under the hood, and how Data Science Actuaries can be using traditional techniques to provide assurance with novel models. Accuracy: how do we evaluate correctness in both discrete and continuous AI outputs – and what’s “good enough”? Confidence: how do we assess the credibility of AI outputs, and know when to escalate to human review? Does AI understand where it doesn’t know it all? Bias: how do we uncover hidden assumptions in language models? Can we measure and mitigate bias to encourage ethical AI use? Actuaries will leave the discussion with a toolkit of validation techniques, rooted in traditional actuarial methods. Attendees will have gain confidence in their ability to manage AI risk, trust AI outputs, and obtain AI transparency.

Hear from Danielle Wood, Chair of the Productivity Commission on Australia's opportunity to reshape its economic trajectory through strategic reform. With productivity growth remaining stubbornly low and impacting living standards, the Government is increasingly turning to its key independent research and advisory body to propose solutions. Danielle will present key findings from the Commission's recent inquiries and outline how their recommendations can drive impactful reform. From workforce participation to regulatory efficiency, technology adoption and service delivery, you'll gain insight into the evidence-based strategies needed to lift Australia's productivity and secure our economic future as multiple sectors go through critical transitions. Discover how actuaries can contribute to this reform agenda — whether through better data utilisation, improved risk management or more effective policy design. Leave with a comprehensive understanding of the productivity challenges ahead and the practical pathways for addressing them.

The boundaries of insurability are shifting across life insurance, superannuation and wealth management. Rising claims costs, changing underwriting paradigms and evolving customer expectations are forcing reassessment of what sustainable protection looks like and how it should be delivered to Australians planning for retirement and beyond. Through a series of lightning talks, this solutions-focused session tackles the ongoing question of insurability in both group and retail products. What have we learned from past missteps? How are insurability parameters changing? What adjustments are needed to ensure protection remains viable and valuable for Australians who need it? Industry leaders will deliver focused insights on the forces reshaping insurability and the practical responses required. Leave with actionable strategies to apply in your own practice across life insurance, superannuation and wealth management. Speakers to be announced soon.

Aged care financing in Australia presents a complex interplay of considerations across individuals, government policy, and aged care providers. For individuals, the financial risks and costs associated with aged care in later life require careful planning and integration with retirement strategies. Financing strategies should integrate superannuation savings and home equity to fund care needs, with growing importance placed on understanding the timing, affordability, and accessibility of services. As longevity increases, so too does the need for sustainable personal financing models that can accommodate retirement income and both residential and in-home care options. From a government perspective, aged care reform is underway with the introduction of a new Aged Care Act and a shift in policy that places greater financial responsibility on individuals, especially in the Support at Home program. This marks a departure from the Aged Care Royal Commission’s recommendation for a dedicated aged care levy, instead favouring increased co-contributions. These changes have significant budgetary implications and highlight a contrast with the National Disability Insurance Scheme (NDIS), which operates under a publicly funded, entitlement-based model. Aged care providers must navigate evolving regulatory and financial landscapes, including prudential standards that affect liquidity, investment practices, and operational margins. The aged care sector faces a wide range of financial sustainability pressures, including rising wages, regulatory compliance demands, and tightening margins. At the same time, with the new Aged Care Act prudential rules, the potential phase-out of RADs, and changes to Support-at-Home pricing and co-payments, providers face challenges in funding future capital requirements, maintaining liquidity, and supporting growth. Providers need to respond with the adoption of AI, workforce innovation, and other strategic financing decisions while balancing care obligations with financial sustainability. Providers—whether for-profit, not-for-profit, or government-operated—face varying pressures in adapting to increased future demand and maintaining service quality. This paper/presentation will explore these intersecting perspectives to illuminate the challenges and opportunities in aged care financing, offering insights into how actuaries can contribute to shaping resilient and equitable solutions for Australia's ageing population.

The life insurance sector stands at a critical juncture within The Great Recalibration - the broader economic and social reset redefining work, wellbeing, and financial security. Amid this transformation, worsening Total and Permanent Disability (TPD) experience and notwithstanding regulatory intervention, Income Protection insurance remains a challenge for the industry. Worsening claims experience has been driven largely by mental health conditions, which has emerged as one of the industry’s most pressing challenges. Beyond its immediate consumer and insurer financial impact, this trend reflects deeper shifts in workforce dynamics, societal expectations, and the role of insurers in supporting long-term resilience. In Australia, mental health now constitutes around one-third of TPD payouts, with insurers disbursing over AUD 2.2 billion in mental health TPD claims in 2024 - nearly double the amount seen five years earlier1 . Further, TPD claim rates for individuals in their 30s attributed to mental health have surged approximately 732 % since 20141 . This trend has significant implications across the insurance value chain - from pricing, underwriting, and claims management to capital and reinsurance strategies. The worsening experience not only impacts profitability and sustainability but also highlights the need for coordinated action across insurers and insurance sectors, reinsurers, regulators, and broader societal stakeholders. This paper examines the underlying drivers of the deterioration in TPD experience, with a particular focus on the rise of mental health-related claims and explores the systemic implications for the life insurance industry. It calls for an industry-wide conversation – one that moves beyond cost management to reimagining how users and distributors of insurance, insurers, reinsurers, and policymakers can collectively respond to the mental health challenge through innovation, partnership, and a renewed focus on societal value.

Employer insolvency poses a significant threat to the integrity of portable long service leave schemes. When businesses collapse, workers may lose access to accrued entitlements, undermining the social contract that these schemes are designed to uphold. In industries like construction, where insolvency risk is elevated due to economic cyclicality and project-based employment, the financial exposure to defaults can be substantial and unpredictable. This paper introduces a credit risk modelling framework to quantify and manage the impact of employer insolvency on leave scheme reserves. Using transition matrices, survival analysis, and industry segmentation, we estimate default probabilities across employer cohorts and simulate the resulting liability shortfalls. The model incorporates historical insolvency data, macroeconomic indicators, and scheme-specific contribution patterns to produce stress-tested reserve adequacy assessments. Our analysis shows that certain employer segments—such as small subcontractors or firms operating in high-volatility regions—pose disproportionate risk to scheme stability. We evaluate the effectiveness of various mitigation strategies, including dynamic reserve buffers, insolvency insurance mechanisms, and early-warning systems based on predictive analytics. These tools enable scheme administrators to proactively manage financial risk and protect worker entitlements even in adverse economic conditions. By applying actuarial techniques to a real-world policy challenge, this paper bridges the gap between financial modelling and social protection. It offers a robust framework for enhancing scheme resilience and aligns with the Actuaries Institute’s commitment to advancing public interest through data-driven decision-making. The insights are directly applicable to regulators, policymakers, and scheme operators seeking to safeguard entitlements in an increasingly uncertain economic landscape.

Background and Motivation The Australian Total and Permanent Disability (TPD) market has recently experienced a significant uplift in claim volumes, driven predominantly by claims related to mental health and musculoskeletal conditions. This surge coincides with a period of economic volatility, where high inflation and rising interest rates have led to a sharp reduction in household disposable income. Given the well-documented link between financial stress and emotional wellbeing, it is highly probable that macro-economic conditions are influencing claim rates. Traditional group pricing techniques are retrospective, relying heavily on historical loss ratios to determine the forward-looking best estimate. However, recent events prompt actuaries to question the completeness of these methods and assess the role the economy should play in informing future pricing exercises. Methodology and Scope This study aims to quantify the correlation between key macroeconomic metrics and Australian group TPD claim rates. Our analysis will focus on identifying a relationship by examining historical downturns, with a particular focus on the 2007-2009 Global Financial Crisis (GFC). The metrics to be tested include the unemployment rate, the consumer confidence index, and changes in real disposable income (amongst others). We will employ time series regression models such as Multiple Linear Regression (MLR) and Autoregressive Integrated Moving Average (ARIMAX) methods to: 1. Establish if a correlation exists. 2. Determine whether this relationship can be reliably modelled, accounting for potential lagged effects (where claims changes follow economic shifts by a set number of months). A successful model will indicate the potential magnitude of claim rate changes driven by economic indicators, allowing for the integration of a prospective element into group pricing methods. Data Challenges and Actuarial Implications A key challenge is the availability of consistent claims data over extended periods. To analyse earlier economic downturns (e.g., the early 1990s recession), we will incorporate proxies for disability claims, such as the Australian Disability Support Pension (DSP) and the US Social Security Disability Insurance (SSDI). It is necessary to account for the interference of external factors, such as past changes to the eligibility criteria of DSP or SSDI policies. Therefore, the modelling will segment the analysis into distinct periods to ensure the correlation is isolated and reliably measured. Ultimately, this research seeks to provide the actuarial community with a more robust and forward-looking framework for forecasting TPD claim rates, improving pricing practices in an increasingly uncertain economic environment.

This presentation raises the curtains on the modelling produced behind the successful public policy campaigns for changes to LISTO and for super to be paid on paid parental leave. Dare I say it Division 296 v2.0 will pass as well and we can say that we are 3 out of 3 on public policy submissions. How did we achieve a 100% success rate? What was the winning formula? What numbers did we produce? Why/how did we select the numbers we produced? How did we use the numbers in storytelling? This presentation focuses on issues in super, but we hope attendees can take the learnings and apply them to driving positive changes in their own practice areas.

Defaults have long shaped outcomes in retirement markets, providing a “safe harbour” for those who may be disengaged or overwhelmed by choice. In Australia, the debate about defaults in retirement drawdown is intensifying as policymakers, funds, and insurers weigh protection against freedom of choice and what we think Australians want. With an ageing population, shifting demographics, and growing pressure on the retirement system, Australia is at a tipping point. The Retirement Income Covenant (RIC), effective since mid-2022, mandated that superannuation trustees develop strategies to assist members in or approaching retirement. But should these strategies include the use of defaults, given the growing importance of lifetime retirement income products? In this session we will present findings from a new survey of Australians, exploring trust in providers, reliance on defaults, and willingness to engage. We will also draw on insights from the UK, US and the Netherlands, where our global presence at Pacific Life Re gives us first-hand knowledge of how defaults and compulsion operate in very different ways with equally different outcomes. By treating defaults as both actuarial levers and indicators of societal values, this session will ask whether Australia’s current settings are fit for the next phase of retirement, and what they reveal about our collective vision of adequacy, choice, and trust.

While digital tools have evolved to support superannuation accumulation, a significant gap exists in providing comprehensive digital advice for the retirement phase. This presentation introduces Aware Super's "Retirement Manager" and explores the actuarial framework that enables scalable retirement income guidance. The retirement phase presents unique complexities that traditional digital solutions struggle to address. Members face critical decisions about how much to withdraw while balancing longevity risk, market volatility, and changing life circumstances. Common problems include drawing too little (leading to unnecessary lifestyle restrictions), drawing too much (risking depletion), and overall lack of confidence in retirement income strategies. Traditional tools focus on individual accounts, but members think at the household level. Retirement Manager addresses these challenges by operating at the household level. It considers both the member and their partner, and incorporates all relevant income sources including the Age Pension, assets outside superannuation, and tax implications. Its hybrid nature enables both direct member self-service and use by Aware Super's advice team to provide emotional validation for members who prefer human assistance. The tool incorporates intelligent nudges including "Maximised to Last" to guide sustainable drawdown strategies, and an inbuilt triage system to identify when human intervention is needed. A key innovation is the Income Confidence Score—a composite actuarial measure that accounts for both the conditional target income shortfall and the risk of superannuation being depleted early. The presentation will explore the actuarial methodology and how we worked within the vendor's projection engine constraints to produce stochastic-like results using a deterministic engine. We'll demonstrate how actuarial concepts are translated into practical member guidance, and share implementation challenges encountered during the development phase, showcasing how actuaries can effectively collaborate with different stakeholders to add value.

As mental health claims surge across Australia’s insurance sector, the issue has become one of the leading drivers of claim costs and market volatility. Current life insurance products often price mental health risks implicitly rather than explicitly—despite mental health now being a dominant cause of claims. This approach increases costs for consumers while destabilising risk pools as claims escalate. This paper presents actuarial evidence to support more data-driven mental health insurance design. Using population-level data from the National Survey of Mental Health and Wellbeing, we quantify transitions between mental health states, recovery patterns, and mortality through Markov-style modelling. By incorporating socio-economic factors such as income, education, and illness persistence, we capture the episodic nature of mental health conditions that static models typically overlook. We identify specific patterns of incidence, persistence, and recovery that enable early intervention strategies, evidence-based pricing approaches, and improved benefit structures. These insights provide actuaries with practical frameworks for balancing commercial sustainability with social equity, positioning organisations to lead rather than react to regulatory and market pressures. Attendees will gain actionable intelligence for product innovation, pricing recalibration, and strategic positioning in a rapidly evolving landscape. This session demonstrates how actuarial rigour can transform mental health insurance from a risk management challenge into an opportunity for market differentiation and meaningful social impact.

Regulators are asking superannuation funds to provide guidance and support to their members as they approach retirement. This paper takes an inventory of the items that need to be considered. It finds that, of the more than 40 items identified, over half arise because regulations are unnecessarily complex and confusing and could be abolished without materially affecting the equity and security of the system. Many of these provisions only benefit the wealthier or better informed. Simplifying the system would create the opportunity to address glaring inequities. Most pressing is that between renters and homeowners. Commonwealth Rent Assistance is universally seen as inadequate for renters. On the other hand, the exemption of the family home from means tests is overgenerous to wealthy homeowners—and distorts the distribution of housing. High effective marginal tax rates on earned income, which arise from the interaction of the welfare and tax systems, are also inequitable and could be addressed. Much of the complexity stems from government efforts to “be seen to act” without considering materiality or side-effects. A simpler system would allow trustees and financial advisors to focus better on the management of the material questions, not least the management of investment and longevity risks. Fairness does not require intricate regulation: effective regulatory design should focus on material issues, maintain logical consistency, and minimise compliance costs for individuals. Actuaries can play a role in maintaining this focus.

How granular should pricing be in the retirement market? The answer is not purely technical, it is also a question of societal goals. Highly differentiated pricing reflects every observable risk and can boost take-up of lifetime income products, as seen in the UK, while pooled pricing spreads risk more broadly for fairness and simplicity, as is common in the Netherlands. With an ageing population and growing pressure on Australia’s retirement system, these choices are increasingly important: how granular should pricing be to encourage engagement, ensure fairness, and support a sustainable retirement market? In this session we will introduce a “fairness frontier” to illustrate the trade-offs and explore where Australia sits today. Drawing on insights from the UK, US, and the Netherlands, where our global presence at Pacific Life Re, gives us first-hand knowledge of market practice, we show how each system balances equity, sustainability, and simplicity in very different ways. By reframing granularity as both a technical and ethical question, this session challenges actuaries to consider how retirement pricing can reflect both commercial realities and member expectations, helping to shape a fairer, more sustainable system for the long term.

There are several widely adopted actuarial systems used by life and general insurers, both in Australia and globally. For example, SAS is commonly used by general insurers, while Prophet is a standard platform for life insurers. These systems have become deeply embedded in organisational processes over many years, providing controls and supporting complex actuarial logic. Insurers have continually maintained and enhanced these models, adding new features and logic to meet evolving product design, pricing, and valuation methodology changes. However, these legacy systems are not without challenges. Their complexity has grown over time, largely due to ongoing customisation required to support a wide array of new products with tailored benefit features. This issue is further exacerbated by the presence of multiple policy administration platforms, each with its own data field specifications. As a result, insurers face environments that are increasingly difficult to control and manage, with inefficiencies and rising maintenance costs. Additionally, the relatively closed nature of these systems restricts integration with modern AI capabilities, and there is a shrinking pool of specialised coders familiar with these languages, making ongoing support and innovation even more challenging. In Australia and globally, more insurers are exploring the use of open-source code—such as Python and Julia—for actuarial modelling. This shift brings a range of benefits, including reduced costs, greater agility in model development, and enhanced transparency and control over code. Tools like automated model-to-open-source code converters allow business users to prototype models in familiar environments and convert them into production-ready code, streamlining requirement gathering and enabling consistent deployment. Real-world examples show that life and general insurers have successfully launched new products, responded to regulatory changes, and improved their actuarial processes by leveraging open-source code solutions. However, migrating to open-source code requires careful planning and consideration. Insurers will need to manage the migration journey, ensure robust controls and auditability, and integrate with existing upstream and downstream systems. Despite these hurdles, the potential gains in cost savings, operational efficiency, and innovation make open-source code an opportunity for insurers to consider as they modernise their actuarial modelling ecosystems.

The aim of this paper is to motivate life insurance actuaries to explore how Bayesian methods can become part of actuarial practice. We will introduce Bayesian approaches, highlighting how they differ from more traditional frequentist methods and why they can be advantageous . We will discuss their flexibility in handling complex models, their natural treatment of uncertainty through posterior distributions, and their ability to incorporate expert judgment or external information via priors. Bayesian methods can allow actuaries to solve actuarial problems in ways that are often difficult under traditional approaches. For example, they allow actuaries to quantify the full distribution of outcomes rather than relying solely on point estimates or asymptotic approximations. They also adapt well to situations with sparse data, shifting business practices, or environments where existing knowledge can add meaningful insights. These strengths can improve decision-making in contexts such as pricing, reserving, and capital modelling, where understanding the range and drivers of uncertainty is critical. Historically Bayesian methods remain underused in life insurance practice. Barriers include the perception that Bayesian models are complex, the high computational burden, and a lack of familiarity among actuaries. However, with the growth of modern statistical software, advances in Markov Chain Monte Carlo and increasing exposure in other actuarial domains such as general insurance, these hurdles are diminishing. We believe that there is opportunity for life actuaries to adopt Bayesian methods more widely. To illustrate this, the paper will present three case studies. The first will explore prediction of incurred but not reported (IBNR) claims, highlighting how prior information and uncertainty quantification can improve insights. The second will examine Bayesian approaches to experience investigations, showing how hierarchical modelling can address sparse data and credibility challenges. Finally, we will demonstrate how Bayesian methods can be applied to group corporate pricing and provide an objective means to address current issues with credibility estimation and distortions in market dynamics.

The Council of Australian Life Insurers (CALI) has warned that the nation’s financial safety net is being pushed to its limits, with more Australians leaving work permanently due to mental ill health than ever before. New data from CALI reveals that mental health is now the leading cause of TPD claims, making up almost one in three claims paid. This presentation considers evidence from claims data, and walks through a TPD case study. It covers: • The implications for underwriting, pricing and product design, linking in the findings from the Actuaries Institute’s Group Insurance workstream of the Disability Insurance Taskforce. • Existing and potential solutions from life insurers and superannuation funds to address sustainability issues. • Parallels to retail product solutions, and implications for the group insurance market.

Since the Royal Commission into Misconduct in the Banking, Superannuation and Financial Services Industry in 2018, litigation in the superannuation sector—including insurance in super—has intensified. Class actions and ASIC enforcement now target practices that may breach core trustee obligations. This paper will explore the landscape. • What were the issues facing recent legal proceedings? • Did governance, systems, or behaviour fail? • What lessons emerge from these failures? • What duties do trustees and insurers owe to members in offering group insurance, especially in areas where consent or disclosure is ambiguous? • Beyond the headline cases, are there latent risks still under the radar in insurance within super? • What role do actuaries have in reducing legal, operational, and reputational risks?

Retirement income systems are under increasing pressure worldwide as populations age, life expectancy rises, and economic conditions shift. Governments, regulators, and individuals face the challenge of ensuring that retirement benefits remain adequate and sustainable. This paper compares Australia’s superannuation and Age Pension systems with Singapore’s Central Provident Fund. Using stochastic modelling, the study projects outcomes to assess the probability of adequacy across the two systems under various scenarios, incorporating investment variability, longevity risk, and the role of housing. It draws lessons on how Australia can learn from Singapore by examining potential changes and their impact on adequacy. Despite differing policy philosophies, economic structures, and cultural expectations, both systems face shared challenges of sustainability, adequacy, and equity as they adapt to ageing populations and evolving economic conditions. The findings provide insights for policymakers, regulators, and actuaries seeking to refine retirement systems or design retirement products that achieve a balance between sustainability, adequacy, and equity.

The introduction of the Retirement Income Covenant (RIC) in July 2022 marked a pivotal shift in Australia’s superannuation landscape, fundamentally transforming how trustees approach retirement income strategy. This paper examines the evolving retirement income solutions offered by superannuation funds in response to the Covenant’s requirements and subsequent regulatory developments. The RIC mandates that superannuation trustees develop, review, and implement a retirement income strategy that assists members in achieving and balancing three key objectives: maximising expected retirement income, managing expected risks to income sustainability, and providing flexible access to savings. This legislative framework has catalysed significant innovation in product design, investment strategies, and member engagement approaches across the industry. Our analysis will explore several emerging trends in how funds are meeting these obligations. Some trustees have enhanced their account-based pension offerings with embedded longevity protection features, including partial annuitisation strategies and dynamic asset allocation frameworks that adjust to member longevity risk. Some growth in the adoption of hybrid products combining guaranteed income streams with market-linked components, allowing members to balance security with growth potential, is expected. The paper explores the actuarial challenges inherent in designing retirement income solutions that satisfy regulatory requirements while remaining commercially viable and member-centric. Key considerations include mortality assumptions, sequencing risk mitigation, means-testing optimisation under Age Pension rules, and the integration of behavioural economics principles to address member inertia and decision-making complexity. We examine how regulatory guidance from APRA and ASIC has shaped implementation approaches, including the requirement for objective-based frameworks and enhanced member communication. The paper also considers the intersection of the RIC with other regulatory developments, including the Quality of Advice Review reforms. Through case studies of fund responses, we will assess the effectiveness of different strategic approaches and identify best practices emerging across the sector. The paper will conclude by considering future directions, including the potential for industry-wide solutions, the role of technology in personalised retirement planning, and outstanding policy questions that may require further regulatory clarification or legislative refinement to optimise member retirement outcomes in an evolving demographic and economic environment.

The life insurance industry stands at the centre of Australia’s economic recalibration. Demographic shifts, longevity trends, climate change, and artificial intelligence are reshaping risk, putting pressure on insurers to adapt models, systems, and ways of working. Life insurers must balance financial sustainability with member protection and trust by proactively transforming how data is used, managed, and governed. Synthetic data offers a key pathway to modernisation, enabling safer, more efficient operations while strengthening actuarial insight and industry resilience. In life insurance, synthetic data enhances pricing, assumption setting, and experience investigations while protecting sensitive policyholder information. AI-driven generative models trained on historical claims data can produce realistic, privacy-preserving datasets that reflect the complex relationships between demographics, product features, and claims outcomes. This approach allows actuaries to work with richer datasets without compromising confidentiality. Synthetic data also helps overcome limited or volatile claim experience, producing more stable pricing patterns across age, occupation, and policy type. It enables robust testing of extreme yet plausible scenarios such as pandemics, climate shocks, and financial crises, strengthening industry resilience and long-term sustainability. Beyond pricing, synthetic data is essential for testing and data migration, particularly as insurers modernise legacy systems and adopt cloud-based, data-driven platforms. It allows full-stage migration testing without exposing customer data, ensuring accuracy, completeness, and system functionality. This reduces the risk of integrity errors, compatibility issues, and costly disruptions while maintaining compliance with data protection standards. Finally, synthetic datasets support scalable, non-disruptive testing and AI-assisted test data management, enabling actuaries and data teams to focus on strategic insights rather than manual validation. By reinforcing ethical data use, privacy compliance, and collaboration across insurers, reinsurers, and regulators, synthetic data empowers the profession to transform uncertainty into opportunity, lift productivity, and deliver both financial and social value in the next economic chapter.

Residential property is the largest asset class in Australia at around $12 trillion. This is nearly triple the size of Australian superannuation balances. Senior Australians have high rates of home ownership. Their homes provide places to live and there are many benefits to home ownership in retirement, both financial and non-financial. But for senior Australians their homes also represent a substantial store of wealth, often untapped. For some, there could be benefits to accessing a component of this wealth.

Home Equity Release

The most obvious way to release housing wealth is to sell the home, but an alternative place to live is then needed. Downsizing works well for a subset of retirees. Then there are “home equity release” product solutions. The most well-known products are reverse mortgages, perhaps better called lifetime loans. These involve borrowing with the home as security, and no repayments until the home is sold. There are also debt-free products and a government scheme.

The Home Equity Access Scheme

This is the (Federal) government scheme, the old Pension Loan Scheme, but with enhancements over the years to make it more flexible and more widely available. It can be a good way to enhance retirement income particularly for those relying solely on the age pension.

Retiree Housing Debt

Not long ago there was a broad public policy presumption that people who own houses retire debt-free. This has shifted with up to 50% of people today retiring with housing debt. What do people with later-life housing debt do? Can they service the debt? Or repay it? If the debt is substantial then perhaps they have to sell their home. Some could take a lump sum from super to pay off debt. An alternative might be to access a portion of their home equity.

As superannuation funds evolve to meet the Retirement Income Covenant and rising member expectations, artificial intelligence offers a promising avenue to enhance retirement guidance and engagement. This presentation explores the concept of an AI-enabled support system that combines predictive analytics with human expertise to deliver timely, personalised nudges and future-self scenarios. Drawing on a holistic view of member data—including transactional, behavioural, and family wealth information—the system aims to anticipate financial shifts and connect members to appropriate support at key decision points. The session will explore the design principles and key features of this AI-human hybrid model, including dynamic net worth analytics, predictive alert mechanisms, and seamless adviser handovers. It will also share early learnings from pilot testing, including member and adviser feedback, and reflect on how such systems can align with broader regulatory expectations around effective communication and member-centric engagement. While not prescriptive, the discussion will touch on how AI-driven nudges and visual storytelling can support principles outlined by Treasury and ASIC, such as clarity, relevance, and timing in retirement communications. By framing AI as a strategic enabler within retirement ecosystems, the presentation invites actuaries to consider how intelligent systems can be responsibly embedded to support better outcomes. It will offer a forward-looking perspective on how predictive intelligence, when paired with human advice, can transform retirement planning from a static process into a dynamic, lifelong journey.

Recent deterioration in total and permanent disability (TPD) claims experience has intensified scrutiny of medical certification processes and portfolio management practices. This presentation examines the key drivers behind these trends and explores targeted hypotheses, with a particular focus on the influence of treating doctors and specialists in certification decisions. We consider how variability in medical opinions, documentation standards and functional assessments can shape claim outcomes, including admission probabilities and dispute frequency.. The session aims to: • Foster discussion on the underlying drivers of TPD experience trends • Utilise case studies and supporting evidence to explore hypotheses • Share perspectives on global practices and potential improvements • Explore innovative approaches to an enduring challenge We conclude with practical considerations for insurers and reinsurers—centred on governance, certification consistency and member-focused measures—to support fair outcomes and portfolio stability.

1. Context & Importance (Why this matters?) Claims experience for the Life insurance industry continues to deteriorate, and claims management is under pressure due to rising complexity, customer expectations, manual and inconsistent process and need for faster cycle times. Actuaries already bring analytical skills through tools like Generalized Linear Models (GLMs) to predict claim severity, cycle time, and risks. And now with Large Language Models LLMs we can process unstructured data that has been trapped in texts from medical notes, claim forms, emails and underwriting notes. Together with GLMs + LLMs, actuaries can support the transformation of claims triage into a faster, smarter, more consistent process and uncover new insights and drivers of claim experience. 2. Approach & Methods (What we will show?) • Simulated Data: Show a representative claims dataset (policyholder attributes, claim types, outcomes) to illustrate how a GLM can rank claims, predicts claim duration, cost, or likelihood of dispute. • Proof of Concept: Show pipeline on how LLMs can extract structured fields from free-text claims descriptions, then feed those into a GLM scoring model. • Conceptual implementation: Show “before vs. after” of the claims process: o Before: manual triage, high cycle times, fraud/misclassification risk, lack of feedback loop of claims management strategies success rate o After: automated triage using GLM scores + LLM text interpretation, support to claims assessors process, improve claims understanding. 3. Expected Outcomes & Implications (So what?) By combining the strengths of GLMs and LLMs, actuaries can move beyond traditional analysis to uncover and quantify a broader set of claims drivers. This integration deepens the understanding of claims experience by capturing both structured and unstructured information, enabling earlier identification of high-risk, complex claims or fraudulent claims. The result is to propose a more efficient and transparent claims management process, with faster triage, better resource allocation, and improved customer outcomes.

Driving a car downhill requires different skills from those required to drive the car uphill. Similarly, to secure optimum superannuation member benefits in the decumulation stage, different asset allocation strategies are required from those which apply in the accumulation stage. WHY? This is primarily due to sequencing risk. Many authors have attempted to define sequencing risk. The definitions are usually reasonably consistent with the definition in this paper. Some have illustrated it with simplified examples, but few, if any, have quantified it and then measured it based on historical data. This research compares the sequencing risk faced by a retiree or an annuitant invested in a balanced portfolio, with that resulting from investing in: • Australian shares, • international shares or • a combination of Australian shares and cash. Comparisons are also included for Australian direct and listed property, semi-government bonds and inflation linked government bonds. In the final section results are presented for ‘real’ sequencing risk allowing for price inflation. The comparisons use both the Austmod historical data and simulated scenarios. The comparisons reveal that Australian shares combined with cash, offers better sequencing risk protection than traditional balanced portfolios. This paper and presentation are an update and extension of an Actuaries Digital article with the same title which was published 30 April 2025: https://www.actuaries.asn.au/research-analysis/sequencing-risk-and-asset-allocation Subsequently, that article was published on Linked-in and since then it has experienced sustained and continuing interest from professional readers.

Accurate mortality prediction underpins actuarial practice, informing insurance pricing, retirement income planning, and product design. Building on recent evidence of socio-economic mortality differentials in Australia (Huang, Hui & Villegas, 2025), this study extends that framework by incorporating health-related variables for retirees and distilling them into a transparent Health Index. Using the ABS Personal Level Integrated Data Asset, linked to Medicare and Pharmaceutical Benefits records, we develop interpretable models that integrate the Health Index as an additional rating factor within the existing socio-economic model. Exploratory analysis shows that mortality risk rises steadily with higher healthcare use, further increased by comorbidity and polypharmacy, but partly offset by regular preventive care. Notably, pain-medication exposure is associated with about threefold higher observed mortality than non-exposed peers, which aligns with the rise in opioid-related deaths over the study window and may reflect addition. A range of machine learning and statistical techniques, including clustering, survival trees, and Markov processes, are employed to identify health-related subgroups and generate life expectancy estimates. By grouping individuals into risk-based cohorts, the approach mitigates adverse selection risks and preserves equity in risk pooling. Preliminary results demonstrate significant gains in accuracy and fairness, with direct implications for fairer retirement outcomes and improved management of longevity risk. To facilitate adoption, the methodology translates complex models into a concise set of health-related questions potentially suitable for underwriting, bridging actuarial modelling and health data analytics. This work complements existing socio-economic mortality research and provides new insights into health-related mortality, offering a foundation for more accurate, transparent, and equitable retirement outcomes in Australia.

With reinsurers holding the risk of about 40% of all life insurance risks in Australia, and having effectively underwritten a lot of the growth in previous decades, it is fair to say that reinsurers are a significant part of the Australian life insurance ecosystem. They have provided access to capital (efficiently) in periods of high growth, services and technical support to aid life insurers grow. But in 2026, the landscape looks very different and so have the factors driving the need for reinsurance. There are fewer – generally larger - life insurers. Most access to directly to capital efficiently from overseas parents, growth is lower reducing the need to access as much additional capital, especially as inforce books are much larger and releasing capital. Furthermore, the life insurers have now built many of the capabilities they used to rely on reinsurance expertise themselves. We already see on large group insurance tenders, more reinsurers participating then life insurers. So, what does this mean for the future of life reinsurance in Australia 10 years from now? This presentation will take a trip down memory lane and demonstrate the role that reinsurance has played in the evolution of life insurance in Australia over the last 20 years. We will then investigate two possible directions of the future state of reinsurance in a decade. • Continue existing models - reinsurance models and coverage similar structures to existing with increasingly transactional-focus and some specialty risk coverage. • Innovate - reinsurance models more boldly adapt and evolve to meet the needs of life insurers in Australia, through exploring different reinsurance structures and/or segments of the market. The existing models aren’t broken and work for those that manage to have enough market share, but are likely to continue to erode margins and lead to a reduction in reinsurers onshore, both in terms of the number and size of teams. But along with innovation, there is a requirement for investment, the appetite to take different risks, and a change management process.

Building true financial wellbeing requires more than strong investment returns. It demands a deeper understanding of human behaviour, mental health and the cognitive factors that influence financial decision-making and retirement confidence.

In this insightful plenary, Dr Fiona Kerr, The NeuroTech Institute, and Alison Bodinnar, Acenda, bring together neuroscience, insurance insights and financial capability frameworks to explore how funds can better support Australians’ financial resilience. The session will highlight evidence-based methods for communicating complex information, improving member confidence and enabling clearer decisions across life stages — ultimately strengthening long-term retirement outcomes.