Events Library

In this insightful webinar, Dr. Małgorzata Śmietanka explores the transformative potential of Large Language Models (LLMs) in the insurance claims process. From automating document OCR and anonymizing sensitive data to reasoning over unstructured text, LLMs offer powerful capabilities to streamline and enhance key claims workflows.

The session focuses on equipping actuaries with a foundational understanding of AI agents, moving beyond buzzwords to actionable insights. The goal is to demystify core concepts, explain key technologies in plain terms, and offer practical starting points for actuarial professionals looking to explore this evolving space. A practical demonstration will showcase how AI agents can support workflows such as data analysis, modeling, and report generation. The session concludes by framing future opportunities for actuaries in the AI ecosystem.

In the era of GenAI, it’s worth emphasizing that material value has been generated already by traditional AI and much potential remains on use cases in future. This session will outline traditional AI success stories, e.g. AI-Augmented underwriting, which has transitioned insurers from pure rules-based underwriting to AI-Augmented underwriting, resulting in improved underwriting automation, customer experience and operational efficiency. By joining the dots between GenAI and traditional AI, a strategic roadmap will be outlined for actuaries and insurers in underwriting and claims. As part of AI-Augmented underwriting, actuarial methods were developed for years to quantify and manage risks of AI models in production.

This paper develops a granular regime-switching framework to model mortality deviations from seasonal baseline trends driven by temperature- and epidemiological-related shocks. The model features three states: (1) a baseline state that captures observed seasonal mortality patterns, (2) an environmental shock state for heat waves, and (3) a respiratory shock state that addresses mortality deviations caused by respiratory outbreaks due to influenza and COVID-19. Transition probabilities between states are modelled using covariate-dependent multinomial logit functions. These functions incorporate, among others, lagged temperature and influenza incidence rates as predictors, allowing dynamic adjustments to evolving shocks.

This paper develops a granular regime-switching framework to model mortality deviations from seasonal baseline trends driven by temperature- and epidemiological-related shocks. The model features three states: (1) a baseline state that captures observed seasonal mortality patterns, (2) an environmental shock state for heat waves, and (3) a respiratory shock state that addresses mortality deviations caused by respiratory outbreaks due to influenza and COVID-19. Transition probabilities between states are modelled using covariate-dependent multinomial logit functions. These functions incorporate, among others, lagged temperature and influenza incidence rates as predictors, allowing dynamic adjustments to evolving shocks. Calibrated on weekly mortality data across 21 French regions and six age groups, the regime-switching framework accounts for spatial and demographic heterogeneity. Under various projection scenarios for temperature and influenza, we quantify uncertainty in mortality forecasts through prediction intervals constructed using an extensive bootstrap approach. These projections can guide insurance companies, healthcare providers, and hospitals in managing risks and planning resources for potential future shocks.