Date of Award

2026

Degree Type

Dissertation

Degree Name

Doctor of Business Administration (DBA)

Department

General Business

First Advisor

Stephen Atlas

Abstract

This study examines Generative Artificial Intelligence (GAI) as both a creator and evaluator of Pay-Per-Click (PPC) advertisements. Drawing on the Elaboration Likelihood Model (ELM) and the Advertising Value Model (AVM), the research examines whether theory-based prompting influences advertising outcomes and whether GAI provides consistent evaluations across repeated assessments. Using a 3 × 10 × 10 experimental design, thirty advertisements were generated under central, peripheral, and baseline prompt conditions and evaluated ten times each, yielding 300 total assessments. Aggregated evaluations demonstrated high reliability, with ICC values exceeding .90, confirming that panel-based aggregation substantially improves consistency relative to single-instance ratings. Results showed that central-route prompts produced advertisements rated highest in informativeness, credibility, and overall advertising value, whereas peripheral prompts increased entertainment but reduced informativeness and credibility. Mediation analyses indicated that informativeness was the only significant pathway linking prompt type to advertising value; entertainment, credibility, and irritation did not significantly mediate this relationship. Overall, the findings suggest that GAI evaluations are more strongly driven by informational and evidence-based content than by affective cues, demonstrating that theory-based prompting enhances the perceived value of AI-generated advertisements and highlighting the potential for structured AI systems to support scalable advertising evaluation and decision-making in marketing contexts.

Download

Available for download on Monday, April 17, 2028

Share

COinS