Objective: With the rapid growth of generative artificial intelligence (AI), AI based content creation has become a dominant trend in contemporary marketing. Although disclosing the use of AI has been recommended as a strategy to promote transparency and ethical accountability, empirical evidence reveals heterogeneous and often negative consumer reactions. To explain this inconsistency, this study adopts a conceptual review approach to identify the processes and factors influencing individuals’ attitudes toward AI generated advertisements.
Method: This research employs a conceptual review methodology through an extensive search in Google Scholar using keywords related to AI disclosure. During the screening process, empirical and conceptual studies examining attitudes, perceptions, and reactions to AI disclosure were included, while studies focused solely on technical aspects of AI without addressing user or consumer responses were excluded. The selected studies were classified into eight core domains. Subsequently, drawing on persuasion knowledge model, the findings were integrated to propose a conceptual framework explaining audience reactions to AI disclosure in advertising, comprising five mediating mechanisms and two moderating variables.
Findings: The proposed conceptual framework suggests that disclosing AI identity tends to have a direct negative effect on audience attitudes, yet this effect operates through several mediating processes. Perceived transparency exerts a positive influence, while inferences of manipulative intent, reduced perceived creativity, algorithm aversion, and decreased perceived legitimacy form the negative pathways. Furthermore, audience experience with AI moderates these effects: positive experience mitigates the negative impact, whereas negative experience amplifies it. Product type also moderates responses, disclosure tends to have weaker negative effects for utilitarian products and stronger ones for hedonic products. The framework’s theoretical contribution lies in consolidating fragmented literature, jointly explaining mediating and moderating mechanisms within a testable model, and extending persuasion knowledge model to the context of intelligent technologies.
Conclusion: This research proposes a set of evidence based practical recommendations for optimizing AI use in advertising content creation. The results indicate that the impact of AI disclosure is neither uniform nor fixed but contingent upon audience perception and contextual factors. Accordingly, when disclosure is mandatory on certain platforms or situations, employing human AI co creation models can be effective. Conversely, in non mandatory contexts, overtly highlighting AI involvement is not always advisable and may be avoided. Disclosure strategies should also align with product type: for utilitarian products, revealing AI use generally has limited negative influence and can be done cautiously; for hedonic products, where emotional, creative, and human elements dominate, AI disclosure may reduce perceived authenticity of the message. In such cases, either avoiding direct disclosure or emphasizing human participation (e.g., implying human collaboration) helps maintain emotional connection and message acceptance. Finally, the study emphasizes the need for clear organizational policies and frameworks for responsible AI utilization. Viewing AI as a tool serving human creativity rather than replacing it allows leveraging technological benefits while preserving authentic human expression, an approach that, in the long run, enhances the effectiveness of marketing communications.