How Identity Types and Intelligence Levels of Digital Humans Affect Consumer Purchase Decisions
Tuotuo Qi, Tianmei Wang, and Jian Tang
International Journal of Electronic Commerce,
Volume 30, Number 3, 2026, pp. 361-393.
Abstract:
Generative AI has accelerated the adoption of hyper-realistic digital humans and digital twins of public figures, while also enhancing their intelligence. However, research on the interaction between digital human identity type and intelligence level remains limited. This study investigates which identity types of digital humans—hyper-realistic digital humans, celebrity digital twins, and expert digital twins—most effectively influence consumer purchase decisions, and how these effects vary across intelligence levels. The results indicate that, relative to hyper-realistic digital humans, celebrity digital twins enhance product recommendation effectiveness primarily through emotional trust, whereas expert digital twins do so through both cognitive and emotional trust. Under low intelligence conditions, hyper-realistic digital humans elicit greater emotional trust, whereas identity type does not significantly affect cognitive trust. In contrast, under high intelligence conditions, expert digital twins generate higher cognitive trust than celebrity digital twins and hyper-realistic digital humans, while expert and celebrity digital twins elicit greater emotional trust than hyper-realistic digital humans. These findings identify optimal digital human identity types and effective identity–intelligence combinations for enhancing consumer trust and purchase decisions, offering both theoretical and practical insights for firms developing digital humans.