Abstract

Despite the rapid rise of generative AI in social media, little is known about how influencer type, AI-curated posts, and disclosure jointly shape user responses. Prior studies show that AI disclosure often reduces trust and authenticity, yet findings remain inconsistent when virtual influencers are involved. Addressing this gap, we propose a model linking influencer type (human vs. virtual) and AI-curated post type (insightful vs. inspirational) to consumer appreciation or aversion toward AI, mediated by perceived authenticity and trust, and moderated by disclosure. Two online experiments (N=24; N=3+) are designed to test these mechanisms. This study contributes to reconciling contradictory findings on AI disclosure in social media and offers practical guidance for managing AI-assisted influencer marketing.

Share

COinS