The Trust Paradox in AI-Driven Customer Support: A Mixed-Methods Analysis of Human vs. AI Trust

Abstract

Following the extensive use of artificial intelligence (AI) by customer care, businesses relyincreasingly on AI-powered chatbots and virtual assistants to enhance customer interaction. However, here is a paradox: do shoppers trust AI customer care more than human agents? This study examines the trust model between customers and AI-powered customer support compared to traditional human agents. We employed a mixed-methods strategy and collected survey responses and sentiment analysis of customer contacts with human and artificial support staff. We measured the key trust indicators, such as response correctness, empathy, efficiency, and user satisfaction. We applied statistical tests to determine significant levels of differences in trust in various customer groups and industries. Our findings reveal that while AI-powered support is perceived as more efficient and consistent, trust varies based on context. Customers tend to trust AI for straightforward queries requiring speed and precision, whereas complex or emotionally sensitive interactions favor human agents due to perceived empathy and understanding. The study also highlights how hybrid AI-human models can bridge trust gaps by leveraging AI’s efficiency with human agents’ emotional intelligence. Future research would have to examine adaptive AI models that enhance contextual comprehension and affective intelligence to build stronger trust. Ethical concerns and AI decision transparency would also have to be considered further in order to enhance AI-based customer interactions’ trust.