The effect of required warmth on consumer acceptance of artificial intelligence in service: The moderating role of AI-human collaboration

Recent technological advances allow artificial intelligence (AI) to perform tasks that require high warmth, such as caring, understanding others' feelings, and being friendly. However, current consumers may be reluctant to accept AI for such tasks. This research investigates the impact of required warmth to conduct a task on consumer acceptance of AI service and the moderating role of AI-human collaboration. A series of choice-based conjoint experiments and one survey yield two main findings. First, consumers tend to refuse AI for tasks that require high warmth due to the low perceived fit between AI and the task at hand. Second, an AI-human collaboration of AI supporting a human employee increases consumer acceptance of AI service for tasks that require high warmth. This is not the case for AI-human collaboration in which AI performs a task that is supervised by a human employee. Theoretically, this study increases our understanding of how consumer acceptance of AI service varies across tasks and how AI-human collaboration can advance AI acceptance. These findings provide insightful suggestions for managers regarding designing AI service and framing AI-human collaboration.