This study explores consumer evaluations of AI-generated advertising, with a specific focus on perceived authenticity. Among the diverse typologies of AI advertising, particular attention is given to machine-dominant content, where AI independently generates the majority of creative elements with minimal human intervention. To highlight the distinct authenticity challenges of this approach, we compare it with human-driven AI advertisements. Using Topic Modeling based on Latent Dirichlet Allocation (LDA) and sentiment analysis, the study examines how consumer responses vary with perceived authenticity and identifies the key dimensions that shape authenticity perceptions. Findings reveal that consumer reactions are deeply rooted in concerns about authenticity, encompassing dimensions such as perceived effort and brand motives. Theoretically, this research extends authenticity scholarship into the domain of machine-generated content. From a managerial perspective, the results offer practical insights for marketers aiming to integrate AI technologies while preserving emotional resonance and trust in brand communications.