Search results

Firms have already begun integrating artificial intelligence (AI) as a replacement for conventional performance management systems owing to its technological superiority. This transition has sparked a growing interest in determining how employees perceive and respond to performance feedback provided by AI as opposed to human supervisors.

A 2 x 2 between-subject experimental design was employed that was manipulated into four experimental conditions: AI algorithms, AI data, highly experienced human supervisors and low-experience human supervisor conditions. A one-way ANOVA and Welch t-test were used to analyze data.

Our findings revealed that with a predefined fixed formula employed for performance feedback, employees exhibited higher levels of trust in AI algorithms, had greater performance expectations and showed stronger intentions to seek performance feedback from AI algorithms than highly experienced human supervisors. Conversely, when performance feedback was provided by human supervisors, even those with less experience, in a discretionary manner, employees' perceptions were higher compared to similar feedback provided by AI data. Moreover, additional analysis findings indicated that combined AI-human performance feedback led to higher levels of employees' perceptions compared to performance feedback solely by AI or humans.

The findings of our study advocate the incorporation of AI in performance management systems and the implementation of AI-human combined feedback approaches as a potential strategy to alleviate the negative perception of employees, thereby increasing firms' return on AI investment.

Our study represents one of the initial endeavors exploring the integration of AI in performance management systems and AI-human collaboration in providing performance feedback to employees.

Although businesses continue to take up artificial intelligence (AI), concerns remain that companies are not realising the full value of their investments. The study aims to provide insights into how AI creates business value by investigating the mediating role of Business Process Management (BPM) capabilities.

The integrative model of IT Business Value was contextualised, and structural equation modelling was applied to validate the proposed serial multiple mediation model using a sample of 448 organisations based in the EU.

The results validate the proposed serial multiple mediation model according to which AI adoption increases organisational performance through decision-making and business process performance. Process automation, organisational learning and process innovation are significant complementary partial mediators, thereby shedding light on how AI creates business value.

In pursuing a complex nomological framework, multiple perspectives on realising business value from AI investments were incorporated. Several moderators presenting complementary organisational resources (e.g. culture, digital maturity, BPM maturity) could be included to identify behaviour in more complex relationships. The ethical and moral issues surrounding AI and its use could also be examined.

The provided insights can help guide organisations towards the most promising AI activities of process automation with AI-enabled decision-making, organisational learning and process innovation to yield business value.

While previous research assumed a moderated relationship, this study extends the growing literature on AI business value by empirically investigating a comprehensive nomological network that links AI adoption to organisational performance in a BPM setting.