Saving face with intelligent unrealistic influencer: How realism of virtual influencer shapes consumer purchase intentions for near-expired products

Purpose While near-expired products are safe to consume, they are often avoided because of social image concerns, which leads to unnecessary waste and undermines sustainability. The purpose of this study is to examine how different types of influencers, including real human influencers, digital human influencers with human-like appearances and intelligent unrealistic influencers with nonhuman appearances, shape consumer responses to near-expired products by investigating whether differences in influencer realism reduce consumers’ perceived face threat and thereby encourage the acceptance of such socially sensitive products. Design/methodology/approach Four experiments were conducted. Study 1 used a 2 (product type: near-expired vs normal) × 3 (influencer type: real human influencers vs digital human influencers vs intelligent unrealistic influencers) between-subjects design. Studies 2A and 2B explored the underlying mechanism through a single-factor design, and Study 3 introduced a boundary condition through a 2 (green label: presence vs absence) × 3 (influencer type) between-subjects design. Findings Across four experiments, this study reveals that compared with real human or digital human influencers, intelligent unrealistic influencers consistently increase purchase intention for near-expired products through reduced perceived face threat. This effect is moderated by the presence of green labels, which reframe the purchase as an environmentally responsible behaviour and offset the face threat of low quality or low cost. Originality/value This research improves the understanding of influencer effectiveness by revealing the psychological mechanism of face threat in socially sensitive contexts and highlights the potential of intelligent unrealistic influencers to promote sustainable consumption.