Generative AI in Public‐Sector Retirement Plans: Behavioral Segmentation Beyond Demographics

Abstract

Generative artificial intelligence (AI), exemplified by conversational tools such as ChatGPT, is rapidly reshaping financial advice delivery, yet its potential within employer‐sponsored retirement plans remains largely unexplored. This study examines the behavioral segments of public‐sector employees most likely to embrace generative AI tools for retirement planning. Primary survey data from 2000 public‐sector employees measured attitudes toward and engagement with AI, resulting in five distinct adopter segments, aligned with Rogers' Diffusion of Innovations theory: Innovators, Early Adopters, Early Majority, Late Majority, and Laggards. Logistic regression analyses revealed significant differences across segments in their interest in four AI‐driven retirement planning functions: income estimation, goal tracking, investment advice, and tax‐efficient withdrawal strategies. Innovators and Early Adopters demonstrated consistently higher interest across all functions, although effect sizes varied by complexity, with the strongest differences observed in investment advice and tax‐efficient withdrawal tools. Traditional demographic factors such as age and income provided inconsistent predictive value compared to behavioral segments, reflecting limitations in demographic‐based targeting strategies. Findings indicate meaningful opportunities for financial professionals and plan sponsors to utilize behavioral segmentation approaches to enhance AI adoption, particularly among employee groups who express interest yet remain underserved by traditional advisory channels. This research provides practical guidance for financial professionals and plan sponsors seeking to identify which employee segments are most receptive to AI‐powered retirement planning tools, with implications for reaching participants who express interest but are not currently served by traditional advisory channels.