Date of Award

4-18-2023

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Computer Science

First Advisor

Elizabeth Pierce

Abstract

Having life insurance is critical for hundreds of millions of people around the world as a key part of sound financial protection for individuals and families, but there are unacceptably high populations that lack coverage. 106 million Americans are uninsured or underinsured, the protection gap being pronounced in communities of color. Making the rigorous process of purchasing insurance easier is crucial in helping protect the financial security of millions. Automated and accelerated underwriting, a relatively nascent practice in the mature life insurance industry, is intended to streamline this process and narrow this coverage gap by making it easier to obtain insurance. The rapid ubiquity of this practice has been driven by the need to improve customer experience, reduce the time required to reach a decision on an application, and provide efficiencies. This process is being enabled by using Artificial Intelligence (AI) and Machine Learning (ML) models that leverage a range of data sources beyond ones that have traditionally been used in underwriting. Being an emerging practice, and absent guidelines, use of these data sources in a safe, transparent, and explainable manner, has been up to individual firms. Long-term success around consumption and use of a myriad of data sources, with more being continually introduced, warrants industry best practices research that is focused on three objectives: 1. Identify external data sources feeding underwriting models across industry in 2022. 2. Provide list to industry, flagging data elements that firms are vigilant about in other industries, and identify companies within Life Insurance that are more mature in their automated underwriting journey. 3. Provide actionable insights and best practices to the industry predicated on measures being undertaken by more mature firms, based on the People, Process, and Technology Framework. This study provides insights that will help firms establish or adjust their automated underwriting practices. The industry case study provides recommendations to ensure that the data being used is of high quality, has appropriate governance, and is free of inadvertent bias and proxy discrimination, so that firms maintain the integrity and efficacy of their ML and AI models, and expand protection for millions of Americans.

Download

Share

COinS