Asset-Liability Management (ALM) sits at the heart of insurance company valuation, risk management, and strategic decision-making. In this two-part web session, we will explore both the foundations and the future of ALM modeling software.
In the first part, participants will discover the fundamental need for ALM models in insurance. We will begin with a review of core accounting concepts—balance sheet, income statement, and cash flow statement—and examine how traditional ALM approaches emerged to link assets and liabilities. The session will then show how modern ALM models transform a book-value balance sheet into an economic one, supporting market-consistent valuation, solvency monitoring, and financial strategy optimization. We will also review the main building blocks of an ALM modeling framework: risk-neutral valuation, economic scenario generation (ESG), asset modeling, liability modeling, and financial and crediting strategy components.
In the second part, we will explore how ALM modeling is evolving in the era of artificial intelligence and advanced technologies. Actuarial models, once limited by deterministic frameworks and static documentation, are now entering a new phase of interactivity, automation, and explainability.
We will discuss how actuaries can leverage modern AI capabilities—such as large language models (LLMs) combined with retrieval-augmented generation (RAG), tool calling, and agentic AI assistants using the Model Context Protocol (MCP)—to fundamentally enhance ALM workflows. These technologies allow models to become context-aware, self-documented, and collaborative, transforming the way teams build, maintain, and interpret ALM projections. Through concrete examples and real-life use cases, participants will discover how AI-driven tools can bring new dimensions to traditional ALM environments:
- Transparency — AI can automatically explain model assumptions, logic, and sensitivities in plain language, improving governance and auditability.
- Interactivity — Analysts can query complex ALM results or scenario outcomes through natural language interfaces, without navigating dense code or reports.
- Intelligence — Virtual assistants can help design simulations, generate test scenarios, or detect inconsistencies between asset and liability models.
Early-bird discount is available for bookings made by 2 February 2026.
