Abstract:

Many enterprises have adopted AI services to respond to customer needs, yet several studies indicate that users are not always satisfied with AI in service. This study explores the key factors that drive users' attitudes toward AI services and usage promotion strategies based on the generation and influence of users' mind perceptions of AI, which including

(1) Mechanism analysis: How do the cues and experiential factors of users' early contact with AI lead to the perception that AI is “good at computing but not feeling”? This study argues that past frustrations with AI in emotional-social service scenarios and mass media portrayals of AI as “highly intelligent and low in emotional intelligence” have led users to develop the perception that AI is good at computation but lacks feelings. This makes users willing to use AI for functional purposes only in cognitive-analytical tasks, but not like to establish empathic connections with AI in emotional-social tasks.

(2) Moderating effects: How do different internal user states and external AI features moderate the formation and activation of this mind perception toward AI? This study believes that higher frequency of AI use and technology use self-efficacy make individuals more likely to establish para-friendship relationships with AI, those two factors that inhibit the formation of mind perceptions that users believe AI is good at computing but lacks feeling. On the other hand, the degree of anthropomorphism and the degree of flexibility in the task influenced individuals' cognitive cues to AI, those two factors that inhibit the activation of the user's mind perception toward AI.

(3) Facilitation strategies: With the reverse utilization of mind perception, how can the mimicry of AI as warmth animals and the provision of technical assistance enable users to accept services from AI in a wider range of scenarios? This study proposes that by designing an information framework that demonstrates the computational power of AI and at the same time, mimics AI as an animal with “high warmth” quality, it can reconcile the paradox of “AI is good at computation but lacks feeling” and promote the user's perception of AI's empathic ability. This will increase users' willingness to use AI in emotional-social tasks. On the other hand, providing users with tips and technical assistance when interacting with AI can help build a joint interaction model between AI and humans, reduce users' concerns about AI's lack of empathic capacity, and increase people's willingness to use AI in emotional-social tasks.

This study attempts to construct a new model of AI service acceptance based on the theory of mind perception at the theoretical level and provide a psychological reference for theoretically explaining users' ambivalent attitudes toward AI services. At the practical level, this study attempts to propose two pathways to facilitate users' acceptance of AI services with the help of mind perception theory and provide a technical reference for enterprises to enhance the effectiveness of AI applications in their service processes.

Key words: artificial intelligence, mind perception, customer service, stereotype, warmth