Flipping the odds of AI-driven open innovation: The effectiveness of partner trustworthiness in counteracting interorganizational knowledge hiding

This paper aims to analyze the negative effect of knowledge hiding on the relationship between artificial intelligence (AI) capability and open innovation (inbound and outbound) when partner trustworthiness (benevolence, integrity, and ability) is high. Structural equations were used to test this three-way interaction with survey data from a sample of 229 firms, mainly from highly digitalized sectors. The findings indicate that interorganizational knowledge hiding affects only the relationship between AI capability and outbound open innovation and that partner ability is the only factor that will counteract this negative effect. Therefore, co-exploitation of AI-based solutions with external allies is the sole scenario to encourage knowledge hiding by increasing employees' perceptions of the likelihood of AI negatively impacting their personal interests at work. Moreover, when trustworthiness is at the forefront of the intraorganizational discussion, the findings downplay the significance of benevolence and integrity as traits that significantly reduce knowledge hiding. In contrast, at the interorganizational level, knowledge hiding is lessened only when employees perceive that co-exploitation with external partners represents an opportunity to learn and capture crucial AI knowledge.