Abstract
As artificial intelligence (AI) and robotic systems increasingly adopt human-like features, organisations face a paradoxical challenge: anthropomorphic design cues simultaneously enhance perceived usefulness while intensifying discomfort, resistance, and societal concerns in human–robot interaction. Prior research has largely examined these effects in isolation, offering limited insight into how users actively reconcile competing beliefs when deciding whether to accept human-like service robots. This paper advances an expectancy–value perspective on anthropomorphism acceptance, proposing that users engage in a deliberative evaluation process in which anticipated service quality is weighed against perceived realistic threats. We conceptualise anthropomorphism as a paradoxical stimulus that inflates both positive and negative expectancies, shaping acceptance through motivated cost–benefit trade-offs. By extending Expectancy–Value Theory (EVT) to human–robot interaction, this developmental paper reframes anthropomorphic robot acceptance as a motivated choice rather than a static attitude, offering a novel theoretical explanation for persistent ambivalence toward anthropomorphism in service settings.