Conversation Analysis of Public-Servant and Persons-of-Interest Face-to-Face Talk via LLM‑Supported AI Script Mapping Tools

Authors

• Arch G. WoodsideFollow
• Suresh Sood, University of Technology SydneyFollow

Document Type

Conference Proceeding

Publication Date

Spring 2026

Abstract

Professional public servants (PSs) including police officers, postal workers, and government tax office and DMV employees interact face-to-face with citizens/persons of interest (POIs), in encounters that are frequently recorded using mobile phones and publicized on platforms such as YouTube.com. Building on our observations about recurrent PS–POI scripts and LLM-supported analysis, we develop five grounded propositions linking enacted scripts to existing theory. The propositions focus on describing early turn-taking asymmetry, the importance of sequential detail beyond-shallow analysis, the predictive role of script recurrence, the value of rich, script-based role-play, and the potential of narrative-intelligence databases for institutional learning. Instead of treating viral PS–POI video clips as episodic scandals, we argue they serve as data-rich laboratories for training, policy design, and research on human–AI conversation analysis in public service contexts.

Copyright

This work is archived and distributed under the repository's Standard Copyright and Reuse License (opens in new tab). End users may copy, store, and distribute this work without restriction. For all other uses, permission must be obtained from the copyright owners or their authorized agents.

DOI

10.20429/amtp.2026.78

Recommended Citation

Woodside, Arch G. and Sood, Suresh, "Conversation Analysis of Public-Servant and Persons-of-Interest Face-to-Face Talk via LLM‑Supported AI Script Mapping Tools" (2026). AMTP Proceedings 2026. 78.
https://digitalcommons.georgiasouthern.edu/amtp-proceedings_2026/78