Background: Mobile applications have emerged as important touchpoints for addressing service requests and optimizing human resources. Within the service industry, the integration of artificial intelligence (AI) into these applications has enabled the inference of product demand, provision of personalized service offers, and enhancement of overall firm value. Customers now engage with these apps to stay informed, seek guidance, and make purchases. It is important to recognize that the interactive and human-like qualities of AI can either foster the co-creation of value with customers or potentially lead to the co-destruction of customer value. Although prior research has examined the process of value co-creation, the present study aims to investigate the underlying factors contributing to the value co-destruction process, specifically within AI-powered mobile applications.

Method: Our research employs topic modelling and content analysis to examine the value co-destruction process that occurs when customers engage with AI apps. We analyze 7,608 negative reviews obtained from eleven AI apps available on Google Play and App Store AI apps.

Results: Our findings reveal six distinct types of value - utilitarian, hedonic, symbolic, social, epistemic, and economic value - that can be co-destroyed during the process. System failure, self-threat and privacy violation are some contributing factors to this value co-destruction process. These values change over time and vary depending on the type of app.

Conclusion: Theoretically, our findings extend the concept of value co-destruction in the context of AI apps. We also offer practical recommendations for designing an AI app in a more service-friendly way.