Artificial intelligence (AI) has become increasingly integrated into digital marketing and social media advertising, enabling companies to produce highly realistic visual content at a low cost and high speed. As AI-generated marketing material becomes more common, concerns regarding authenticity, trust and transparency have become more relevant for both consumers and companies. This study investigates how consumers perceive AI-generated versus human-made marketing content in social media format and whether disclosure conditions affect consumers’ evaluations.

A quantitative experimental design using A/B testing was conducted with 40 participants who collectively evaluated 400 fictional social media advertisements under three different disclosure conditions: Full disclosure, Partial disclosure and No disclosure. The participants rated ten pieces of content each based on aesthetic appeal, trustworthiness, dependability, professionalism and interest in the product. Out of these ten advertisements, five were generated fully with AI and five were human-made with stock photography and Adobe Photoshop.

The findings indicate that human-made advertisements received higher appraisal regarding trustworthiness and dependability, whilst no significant differences were identified for the aesthetic appeal. The participants in this study also demonstrated a relatively high accuracy in detecting and identifying AI-generated images. Additionally, perceived AI involvement influenced evaluations more strongly than the actual production method, suggesting that the consumer's skepticism toward marketing is primarily psychological rather than visual.

This study contributes to the growing body of research on AI-generated marketing communication by highlighting the importance of authenticity, transparency and consumer trust in an increasingly AI-assisted advertisement environment.