The integration of Large Language Models (LLMs) into insurance operations is expected to transform traditional insurance claims handling. Despite anecdotal evidence from single cases by insurance providers and agencies, a systematic assessment of the performance and potential biases of LLMs to perform claims handling across a wide variety of insurance domains is absent. This paper presents a systematic evaluation of LLM-powered claims handling across three studies. First, we benchmark LLM performance against human insurance clerks using standardized exam cases. Second, we assess the performance and potential biases of LLM-based claims handling in simulated cases across four major insurance domains and tasks. Third, we examine the applicability of LLMs by comparing their claim assessments to those of an expert third-party administrator based on real-world claim cases. Our findings show that LLMs process claims not only efficiently (seconds of processing image data, processing and matching text data from policy documents and damage reports, etc.) but also highly accurately (close to perfect assessments for simple claims and up to 94% accuracy for more complex claims involving multiple parties and policies), often exceeding human benchmarks and reducing claims leakage. These findings have important implications for the future of automated claims handling and the potential to not only substantially improve the speed but also the accuracy of claims handling in industry practice.

This short case covers the recent strategic developments of Barnes & Noble, the largest US book chain, and puts them in the context of its peers worldwide (in the UK, Canada, Germany and France). The perspective of the case is the one of James Daunt, CEO of Barnes & Noble since August 2019. Readers are put in his shoes and are asked to analyze Barnes & Noble's response strategy to an increasingly digital book market. This short case is intended to serve as part of a course session on digitalization strategies or strategic change responding to macro trends. The case can support a full 45-minute class session, or can be integrated into a 90-minute session that also includes theory input by the lecturer. The case questions are open-ended but intended to introduce frameworks to assess strategic change and response strategies to exogenous shocks.