Nontechnical and technical artificial intelligence capability: Their antecedents and impacts on firm performance

Abstract

Artificial intelligence (AI) has attracted huge attention in management field. Its application in organizations has become a common phenomenon. Academics are actively studying how to promote firms to use AI techniques more effectively and the impact of this phenomenon. Especially based on the resource-based view (RBV), scholars have investigated the resources and capabilities that are helpful for firms to apply AI, and developed relevant concepts such as AI capability (AIC). However, there are still many issues that have not been studied, for example what factors can facilitate the improvement of AIC, what factors affect the relationship between AIC and organizational performance, etc. In order to fill these research gaps, this study proposes organizational and contextual antecedents that may influence the development of firm AIC based on RBV and institutional theory. Through the review of AI research, the concepts of nontechnical AIC (NAIC) and technical AIC (TAIC) are constructed from the perceived divergence of nontechnical and technical research. It also proposes corresponding conceptual model and empirically tests the relationships between NAIC and TAIC and different antecedents, as well as how they ultimately affect firm performance. The data was collected from 206 firms in the Yangtze River Delta region of China that have used AI techniques for more than a year. SPSS is used to perform structural equation model analysis and test hypotheses. Data analysis results show that exploitation strategy, coercive pressure, and mimetic pressure can improve firm NAIC. Exploration, leaders’ AI knowledge, and mimetic pressure will improve firm TAIC, and these relationships are moderated to varying degrees by the firm’s data-driven culture. NAIC and TAIC both have a very significant positive impact on firm performance, and they are also moderated by firm international presence. These findings confirm the feasibility of understanding and studying AIC from the perspective of technical relevance, make theoretical contributions to AI-related research, and provide suggestions for management practices of firms applying AI techniques.