Date of Completion

Spring 5-4-2026

Thesis Advisor(s)

Dan Watt; Britta Hay

Honors Major

Mathematics/Actuarial Science/Finance

Abstract

Artificial intelligence (AI) is rapidly transforming the property and casualty (P&C) insurance industry by changing how insurers assess risk, price policies, detect fraud, process claims, and interact with customers. This paper examines the evolution of AI from machine learning to generative AI and agentic AI and evaluates the impact of these technologies on core insurance functions. Through a review of academic literature, industry research, and current insurance applications, the paper analyzes both the opportunities and challenges associated with AI adoption in P&C insurance.

The findings indicate that AI improves operational efficiency, underwriting accuracy, pricing precision, fraud detection capabilities, claims processing speed, and customer service responsiveness. AI systems can process large volumes of structured and unstructured data, identify patterns that may be overlooked by human professionals, and automate routine decision-making processes. However, significant limitations remain, including model opacity, data privacy concerns, algorithmic bias, false positives in fraud detection, and an inability to fully understand complex contextual, ethical, and regulatory considerations.

The paper concludes that while AI will continue to reshape insurance operations and increase automation, it is unlikely to fully replace human judgment. High-stakes decisions involving fairness, ambiguity, regulatory compliance, and customer relationships require human expertise and oversight. As the industry evolves toward greater automation and data-driven decision-making, the most effective model will be a hybrid approach that combines AI’s analytical capabilities with human intervention, enabling insurers to improve efficiency and customer outcomes while maintaining accountability, transparency, and ethical decision-making.

Accessibility Requirements

1

Download

Share

COinS