Title: Modeling Online Public Discourse on AI-Generated Digital Magazine Content: An Exploratory Regression-Based Study

This study examines online public discourse surrounding AI-generated digital magazine content using a manually curated dataset of 125 Reddit comments from controversies involving AI-generated fashion imagery, fabricated editorial interviews, and broader debates over the use of AI in editorial production. Each comment was coded on a three-point scale for sentiment and also for the presence of six thematic concerns that are authenticity, trust, creativity, labor displacement, beauty standards, and pragmatic AI use. To test how these themes were linked to sentiment, a simple linear regression model was curated and used. A computational text-classification procedure was also applied as a layer of validation to strengthen the methodological transparency of the coding framework, assessing whether the manually coded thematic variables could be recovered from comment text. The dataset indicated that discourse was predominantly negative within the sample that 71.2% of comments were negative, followed by 18.4% for neutral comments and 10.4% for positive comments. 57.9% of the variance in sentiment was explained by the regression model. The strongest positive predictor of sentiment was pragmatic AI use, while beauty-standard framing was also positively linked to sentiment. Authenticity, trust, and labor concerns were all directionally negative but were not significantly independent in the full model. Overall, the findings suggest that how AI was framed as a deceptive replacement for human creativity or a supportive but limited production tool played a more significant role in shaping the sentiment within this curated Reddit discourse sample more than AI alone. This study offers an exploratory framework for analyzing online discourse about AI-mediated editorial media.